DETAILED ACTION
Notice of AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
Applicant’s Amendment and remarks dated 5/5/2026 have been considered. Claims 13-20 have been added. Claims 1-20 are pending.
Claim Objections. The objections to claims 4 and 5 are withdrawn in view of Applicant’s amendments to such claims.
35 U.S.C. 112(b) Rejections. The rejections to claims 1-2 under 35 U.S.C. 112(b) are withdrawn in view of Applicant’s amendments to the claims to fix the incorrect Markush grouping.
Response to Arguments
On page 11 of Applicant’s 5/5/2026 Amendment and remarks, Applicant asserts that no new matter has been added via this amendment.
The examiner respectfully disagrees. See 35 U.S.C. 112(a) rejections below.
On page 13 of Applicant’s 5/5/2026 Amendment and remarks, with respect to the rejection of claim 1 under 35 U.S.C. 101, with respect to Step 2A, Prong 1, Applicant argues:
PNG
media_image1.png
182
661
media_image1.png
Greyscale
The examiner respectfully disagrees. Applicant has not rebutted the examiners findings that several of the limitations of claim 1 recite mental processes.
On pages 13-14 of Applicant’s 5/5/2026 Amendment and remarks, with respect to the rejection of claim 1 under 35 U.S.C. 101, with respect to Step 2A, Prong 2, Applicant argues that claim 1 “reflects a technical improvement to the functioning of an aircraft data acquisition and processing system, rather than a mere abstract data analysis.” Applicant further argues:
PNG
media_image2.png
474
648
media_image2.png
Greyscale
The examiner respectfully disagrees. The claims merely recite generic machine learning techniques in the field of use of estimating aircraft weight. No improvements have been made to any machine learning technologies. If anything, the only improvements are to the mental processes of estimating aircraft weight. Patent claims that do no more than claim the application of generic machine learning to new data environments, without disclosing improvements to the machine learning models to be applied, are patent ineligible under Section 101.
On page 14 of Applicant’s 5/5/2026 Amendment and remarks, with respect to the rejection of claim 1 under 35 U.S.C. 101, with respect to Step 2B Applicant argues:
PNG
media_image3.png
284
654
media_image3.png
Greyscale
The examiner respectfully disagrees. MPEP 2106.05(d) explains that the “well-understood, routine, conventional activities” consideration is not a standalone test for eligibility. Because Applicant provides no actual evidence that the claims recite activities that are not “well-understood, routine, conventional activities”, the examiner finds that this consideration does not favor, nor disfavor, a finding of eligibility.
On pages 15-18 of Applicant’s 5/5/2026 Amendment and remarks, with respect to the rejection of claim 1 under 35 U.S.C. 103, Applicant provides pages of arguments about what MCCOOL and MORALES allegedly teach, and then concludes that:
PNG
media_image4.png
118
656
media_image4.png
Greyscale
The examiner respectfully disagrees. Pages 28-29 of the 2/5/2026 office action specifically set forth how the combination of MCCOOL and MORALES teaches the “if the at least one consistency test is validated, storing the at least one set of flight data and the weight(s) calculated at the point in time t, each stored set of flight data associated with a calculated weight forming a set of training data for the at least one machine learning artificial intelligence model” limitation, and Applicant has not rebutted any of the examiner’s findings in its long-winded response. Applicant’s response ignores col. 12, lines 33-42 of MCCOOL, which explain how measurements are stored in a knowledge base, where such measurements from sensors correspond to the recited “flight data”, and which in combination with the teachings of MORALES, are now only stored together with the weight of the aircraft if certain measurements are compared to a threshold to determine if the measurement is consistent with previous measurements.
On page 19 of Applicant’s 5/5/2026 Amendment and remarks, with respect to the rejection of claim 1 under 35 U.S.C. 103, Applicant writes:
PNG
media_image5.png
376
650
media_image5.png
Greyscale
Applicant’s summary of what Applicant contends claim 1 covers does not address the actual language of the claims.
On page 19 of Applicant’s 5/5/2026 Amendment and remarks, with respect to the rejection of dependent claims 2-12 under 35 U.S.C. 103, Applicant argues that such claims should be allowed for the same reasons argued with respect to claim 1.
The examiner respectfully disagrees for the same reasons explained above with respect to claim 1.
On page 19 of Applicant’s 5/5/2026 Amendment and remarks, with respect to new claims 13-20, Applicant argues that such claims corresponds to claims 1-5 and 9-11, and should be allowed for the same reasons.
The examiner notes that while there are similarities, new claims 13-20 have appreciable differences from claims 1-5 and 9-11. Regardless, such claims are all rejected under both 35 U.S.C. 101 and 103 as set forth herein.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
Claims 1-20 are rejected under 35 U.S.C. 112(a) as failing to comply with the written description requirement. The claims contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, at the time the application was filed, had possession of the claimed invention.
Independent claim 1 recites:
wherein sets of flight data that do not satisfy the consistency test are excluded from the training data so as to improve reliability of the training data used by the machine learning artificial intelligence model
Independent claim 13 recites:
exclude sets of flight data that do not satisfy the consistency test from the training data so as to improve reliability of the training data
MPEP 2163 I.B explains: “While there is no in haec verba requirement, newly added claims or claim limitations must be supported in the specification through express, implicit, or inherent disclosure.”
While paras. 0102 and 0103 of the instant specification disclose that sets of flight data J1, J2 can be used to form a set of training data if one of the consistency tests are validated, the instant specification is silent as to whether the exclusion of data that does not satisfy the consistency test will improve reliability of the training data as recited in the amendments to claim 1 and new claim 13. Therefore, there is no explicit support in the specification for the above-mentioned limitations in claim 1 and 13.
The examiner also finds that there is no implicit or inherent disclosure in the specification to support the claim limitation that excluding sets of flight data that do not satisfy the consistency test from the training data will improve reliability of the training data. Indeed, one of ordinary skill in the art would understand that having negative test data (e.g., data known and labeled as failing the consistency test) would also be beneficial to training machine learning models, so it is neither implicit nor inherent that removing such negative test data from the training set would improve the reliability of the training data.
Claims 2-12 depend from claim 1 and do not remedy the deficiencies of claim 1, and are therefore rejected for the same reasons explained with respect to claim 1.
Claims 14-20 depend from claim 13 and do not remedy the deficiencies of claim 31, and are therefore rejected for the same reasons explained with respect to claim 13.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding Step 1 of the Alice/Mayo framework, Claims 1-12 are directed to a method (a process), and claims 13-20 are directed to an aircraft (a machine) which are each one of the four statutory categories of inventions.
Regarding Claim 1
Step 2A, prong 1 (Is the claim directed to a law of nature, a natural phenomenon or an abstract idea).
Claim 1 recites the following mental processes, that in each case under the broadest reasonable interpretation, covers performance of the limitation in the mind (including an observation, evaluation, judgment, opinion) or with the aid of pencil and paper but for the recitation of generic computer components (e.g., “machine learning artificial intelligence model”, “controller”, “memory”).
the method improving reliability of flight data used for training and thereby improving estimation of aircraft weight during operation (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, e.g., by only selecting reliable flight data to train a neural network model, thus improving the output of estimation of aircraft weight during operation by selecting only reliable flight data)
carrying out at least one predetermined consistency test..., the at least one consistency test configured to to check that a reliable reference weight has been calculated or is capable of being calculated; (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by comparing a measured weight to a reliable reference weight to determine if the measured weight is capable of being reliably calculated)
calculating, ... at least one calculated weight of the aircraft as a function of at least one weight chosen from the group consisting of a current boarded fuel weight, a consumed fuel weight, a previously measured weight of the aircraft, an empty weight of the aircraft, a payload previously input by a user, a weight measured by a piece of connected equipment, and the reference weight; (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by using a mathematical calculation, using the recited parameters as dependent variables, to calculate a predicted weight of the aircraft)
wherein sets of flight data that do not satisfy the consistency test are excluded from the training data so as to improve reliability of the training data used by the machine learning artificial intelligence model (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, e.g., by excluding unreliable flight data from being used to train a neural network model, thus improving the output of estimation of aircraft weight during operation by selecting only reliable flight data)
Step 2A, prong 2 (Does the claim recite additional elements that integrate the judicial exception into a practical application?).
The judicial exception is not integrated into a practical application. In particular, the claim recites the additional elements (e.g., “machine learning artificial intelligence model”, “controller”, “memory”) which are recited at a high-level of generality such that they amount to no more than mere instructions to apply the exception using a generic computer component (See MPEP 2106.05(f)).
Regarding the “A method for training at least one machine learning artificial intelligence model” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of training a ML AI model. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (a training a ML AI model). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Regarding the “the at least one machine learning artificial intelligence model being configured to be stored in a memory of an aircraft or a ground station” limitation, such additional element of a data storage step is recited at a high level of generality and amounts to extra-solution activity of storing data, i.e. post-solution activity of data storage for use in the claimed process (see MPEP 2106.05(g)).
Regarding the “configured to be implemented during at least one predetermined flight phase of at least one aircraft of the same type” limitation, such limitation amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use. As explained by the Supreme Court, a claim directed to a judicial exception cannot be made eligible "simply by having the applicant acquiesce to limiting the reach of the patent for the formula to a particular technological use." Diamond v. Diehr, 450 U.S. 175, 192 n.14, 209 USPQ 1, 10 n. 14 (1981). Thus, limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception do not integrate a judicial exception into a practical application.
Regarding the “the method comprising carrying out a plurality of flights, and, for at least one of the plurality of flights, acquiring, during flight, at least one set of flight data acquired with several sensing devices at the same point in time t, wherein, for the plurality of flights” limitation, such additional element of a data gathering step is recited at a high level of generality and amounts to extra-solution activity of receiving data, i.e. pre-solution activity of gathering data for use in the claimed process (see MPEP 2106.05(g)).
Regarding the “with at least one consistency controller” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of a controller. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (a controller). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Regarding the “with at least one weight calculation controller” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of a controller. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (a controller). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Regarding the “if the at least one consistency test is validated, storing the at least one set of flight data and the at least one calculated weight at the point in time t, each stored set of flight data associated with a calculated weight forming a set of training data for the at least one machine learning artificial intelligence model” limitation, such additional element of a data storage step is recited at a high level of generality and amounts to extra-solution activity of storing data, i.e. post-solution activity of data storage for use in the claimed process (see MPEP 2106.05(g)).
Regarding the “using the sets of training data to program the at least one machine learning artificial intelligence model, the at least one machine learning artificial intelligence model being configured to estimate, at any time, an estimated instantaneous weight of the aircraft, or another aircraft of the same type, based on a current set of flight data” limitation, such limitation amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (predicting the weight of an aircraft). As explained by the Supreme Court, a claim directed to a judicial exception cannot be made eligible "simply by having the applicant acquiesce to limiting the reach of the patent for the formula to a particular technological use." Diamond v. Diehr, 450 U.S. 175, 192 n.14, 209 USPQ 1, 10 n. 14 (1981). Thus, limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception do not integrate a judicial exception into a practical application.
Step 2B (Does the claim recite additional elements that amount to significantly more than the judicial exception?)
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception. As discussed above, the additional elements (e.g., “machine learning artificial intelligence model”, “controller”, “memory”) are recited at a high-level of generality such that they amount to no more than mere instructions to apply the exception using a generic computer component (See MPEP 2106.05(f)).
Regarding the “A method for training at least one machine learning artificial intelligence model” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Regarding the “the at least one machine learning artificial intelligence model being configured to be stored in a memory of an aircraft or a ground station” limitation, as discussed above, the additional element of a data storage step is recited at a high level of generality and amounts to extra-solution activity of storing data, i.e. post-solution activity of storing data after or during use in the claimed process. The courts have found limitations directed to storing information electronically, recited at a high level of generality, to be well-understood, routine, and conventional (see MPEP 2106.05(d)(II), "electronic record keeping," and "storing and retrieving information in memory").
Regarding the “configured to be implemented during at least one predetermined flight phase of at least one aircraft of the same type” limitation, such limitation amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use as explained above, which does not amount to significantly more than the judicial exception. MPEP 2106.05(h).
Regarding the “the method comprising carrying out a plurality of flights, and, for at least one of the plurality of flights, acquiring, during flight, at least one set of flight data acquired with several sensing devices at the same point in time t, wherein, for the plurality of flights” limitation, as discussed above, the additional element of a data gathering step is recited at a high level of generality and amounts to extra-solution activity of receiving data, i.e. pre-solution activity of gathering data for use in the claimed process. The courts have found limitations directed to obtaining information electronically, recited at a high level of generality, to be well-understood, routine, and conventional (see MPEP 2106.05(d)(II), “receiving or transmitting data over a network”, "electronic record keeping," and "storing and retrieving information in memory").
Regarding the “with at least one consistency controller” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Regarding the “with at least one weight calculation controller” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Regarding the “if the at least one consistency test is validated, storing the at least one set of flight data and the at least one calculated weight at the point in time t, each stored set of flight data associated with a calculated weight forming a set of training data for the at least one machine learning artificial intelligence model” limitation, as discussed above, the additional element of a data storage step is recited at a high level of generality and amounts to extra-solution activity of storing data, i.e. post-solution activity of storing data after or during use in the claimed process. The courts have found limitations directed to storing information electronically, recited at a high level of generality, to be well-understood, routine, and conventional (see MPEP 2106.05(d)(II), "electronic record keeping," and "storing and retrieving information in memory").
Regarding the “using the sets of training data to program the at least one machine learning artificial intelligence model, the at least one machine learning artificial intelligence model being configured to estimate, at any time, an estimated instantaneous weight of the aircraft, or another aircraft of the same type, based on a current set of flight data” limitation, such limitation amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use as explained above, which does not amount to significantly more than the judicial exception. MPEP 2106.05(h).
Regarding Claim 2
Step 2A, Prong 1
wherein the reference weight is defined as a function of an estimated weight of the aircraft based on measurements from several sensors, each sensor being arranged on each landing gear of the aircraft, (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by defining a reference weight so that it is determined based on a function of measurement inputs as recited)
the at least one consistency test comprising calculating a difference between the calculated weight and the estimated weight and then checking that the difference is less than a difference threshold value. (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by calculating a difference between the calculated and estimated weight and then comparing that difference to a threshold)
Regarding Step 2A, Prong 2, the claim does not include any additional elements that integrate the judicial exception into a practical application and regarding Step 2B, there are no additional elements recited that amount to significantly more than the judicial exception.
Regarding Claim 3
Step 2A, Prong 1
wherein the reference weight is defined as a function of a checked theoretical weight of the aircraft, (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by defining a reference weight so that it is determined based on a function of a checked theoretical weight)
the at least one consistency test comprising checking the parameterization of the checked theoretical weight. (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by checking the different parameters used to determine the checked theoretical weight)
Regarding Step 2A, Prong 2, the claim does not include any additional elements that integrate the judicial exception into a practical application and regarding Step 2B, there are no additional elements recited that amount to significantly more than the judicial exception.
Regarding Claim 4
Step 2A, Prong 1
the reference weight is defined as a function of a calculated theoretical weight of the aircraft, (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by defining a reference weight so that it is determined based on a function of a calculated theoretical weight)
the at least one consistency test comprising checking that the at least one piece of information has been parameterized. (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by checking the different parameters used to determine the calculated theoretical weight)
Step 2A, Prong 2
Regarding the “the calculated theoretical weight being obtained by the at least one weight calculation controller based on at least one piece of information parameterized with a human-machine interface of the aircraft by the user prior to the at least one consistency test” limitation, such additional element of a data gathering step is recited at a high level of generality and amounts to extra-solution activity of receiving data, i.e. pre-solution activity of gathering data for use in the claimed process (see MPEP 2106.05(g)).
Step 2B
Regarding the “the calculated theoretical weight being obtained by the at least one weight calculation controller based on at least one piece of information parameterized with a human-machine interface of the aircraft by the user prior to the at least one consistency test” limitation, as discussed above, the additional element of a data gathering step is recited at a high level of generality and amounts to extra-solution activity of receiving data, i.e. pre-solution activity of gathering data for use in the claimed process. The courts have found limitations directed to obtaining information electronically, recited at a high level of generality, to be well-understood, routine, and conventional (see MPEP 2106.05(d)(II), “receiving or transmitting data over a network”, "electronic record keeping," and "storing and retrieving information in memory").
Regarding Claim 5
Step 2A, Prong 2
Regarding the “wherein the flight data has data relating to at least two flight parameters chosen from the group consisting of a speed of the aircraft in relation to the air, a vertical speed of the aircraft in relation to the ground, a longitudinal speed of the aircraft in relation to the ground, a lateral speed of the aircraft in relation to the ground, a vertical acceleration of the aircraft in relation to the ground, a flow of fuel supplying an engine of the aircraft, a rotational speed of a rotor equipping the aircraft, a wind direction, a wind speed, a quantity of fuel on board the aircraft, a yaw trajectory of the aircraft, an attitude of the aircraft, an air density, an air temperature, an altitude of the aircraft, an angle of attack of a wing of the aircraft, a power consumed by at least one engine of the aircraft, positions of flight controls and positions of blades of a rotor and/or of a propeller” limitation, this limitation merely describes the types of data used to estimate aircraft weight, and therefore such limitation amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use. As explained by the Supreme Court, a claim directed to a judicial exception cannot be made eligible "simply by having the applicant acquiesce to limiting the reach of the patent for the formula to a particular technological use." Diamond v. Diehr, 450 U.S. 175, 192 n.14, 209 USPQ 1, 10 n. 14 (1981). Thus, limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception do not integrate a judicial exception into a practical application.
Step 2B
Regarding the “wherein the flight data has data relating to at least two flight parameters chosen from the group consisting of a speed of the aircraft in relation to the air, a vertical speed of the aircraft in relation to the ground, a longitudinal speed of the aircraft in relation to the ground, a lateral speed of the aircraft in relation to the ground, a vertical acceleration of the aircraft in relation to the ground, a flow of fuel supplying an engine of the aircraft, a rotational speed of a rotor equipping the aircraft, a wind direction, a wind speed, a quantity of fuel on board the aircraft, a yaw trajectory of the aircraft, an attitude of the aircraft, an air density, an air temperature, an altitude of the aircraft, an angle of attack of a wing of the aircraft, a power consumed by at least one engine of the aircraft, positions of flight controls and positions of blades of a rotor and/or of a propeller” limitation, such limitation amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use as explained above, which does not amount to significantly more than the judicial exception. MPEP 2106.05(h).
Regarding Claim 6
Step 2A, Prong 1
wherein, prior to the use of the sets of training data, the method comprises a count for counting the number N of the sets of training data and a comparison between the number N and a predetermined threshold value S. (under the broadest reasonable interpretation, a human such as a computer scientist can perform this step mentally, such as by counting the number of sets of training data and comparing that number to a predetermined threshold)
Regarding Step 2A, Prong 2, the claim does not include any additional elements that integrate the judicial exception into a practical application and regarding Step 2B, there are no additional elements recited that amount to significantly more than the judicial exception.
Regarding Claim 7
Step 2A, Prong 1
wherein the use of the sets of training data is implemented when the number N is greater than or equal to the predetermined threshold value S. (under the broadest reasonable interpretation, a human such as a computer scientist can perform this step mentally, such as by counting the number of sets of training data and comparing that number to a predetermined threshold, and only when the threshold is exceeded, making the decision to use the sets of training data)
Regarding Step 2A, Prong 2, the claim does not include any additional elements that integrate the judicial exception into a practical application and regarding Step 2B, there are no additional elements recited that amount to significantly more than the judicial exception.
Regarding Claim 8
Step 2A, Prong 2
Regarding the “wherein the at least one machine learning artificial intelligence model comprises a first model and a second model different from the first model, the first model being associated with a first predetermined flight phase from the at least one predetermined flight phase and the second model being associated with a second predetermined flight phase from the at least one predetermined flight phase, the first predetermined flight phase being different from the second predetermined flight phase” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of dividing a model into sub-models. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (machine learning sub-models). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Step 2B
Regarding the “wherein the at least one machine learning artificial intelligence model comprises a first model and a second model different from the first model, the first model being associated with a first predetermined flight phase from the at least one predetermined flight phase and the second model being associated with a second predetermined flight phase from the at least one predetermined flight phase, the first predetermined flight phase being different from the second predetermined flight phase” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Regarding Claim 9
Step 2A, Prong 1
wherein the at least one predetermined flight phase is chosen as a function of a required accuracy of the at least one machine learning artificial intelligence model for estimating, at any time, the estimated instantaneous weight of the aircraft, or an aircraft of the same type, based on a current set of flight data. (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by selecting a flight phase to prioritize accuracy of the machine learning model)
Step 2A, Prong 2
Regarding the “wherein the at least one machine learning artificial intelligence model is applied during the chosen predetermined flight phase to estimate the estimated instantaneous weight based on the current set of flight data” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of using a machine learning model to estimate an output based on certain input. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (machine learning models). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Step 2B
Regarding the “wherein the at least one machine learning artificial intelligence model is applied during the chosen predetermined flight phase to estimate the estimated instantaneous weight based on the current set of flight data” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Regarding Claim 10
Step 2A, Prong 1
wherein the at least one predetermined flight phase is chosen as a function of a required dispersion of the at least one machine learning artificial intelligence model for estimating, at any time, the estimated instantaneous weight of the aircraft, or an aircraft of the same type, based on a current set of flight data. (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by selecting a flight phase out of a distribution (or dispersion) of different flight phases, to pick the closest flight phase for the most accurate measurements)
Step 2A, Prong 2
Regarding the “wherein the at least one machine learning artificial intelligence model is applied during the chosen predetermined flight phase to estimate the estimated instantaneous weight based on the current set of flight data” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of using a machine learning model to estimate an output based on certain input. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (machine learning models). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Step 2B
Regarding the “wherein the at least one machine learning artificial intelligence model is applied during the chosen predetermined flight phase to estimate the estimated instantaneous weight based on the current set of flight data” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Regarding Claim 11
Step 2A, Prong 1
wherein the at least one predetermined flight phase is chosen as a function of a diversity of the plurality of flights performed by the user of the aircraft, or an aircraft of the same type. (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by selecting a flight phase out of a distribution (or diversity) of different flight phases, to pick the closest flight phase for the most accurate measurements)
Regarding Step 2A, Prong 2, the claim does not include any additional elements that integrate the judicial exception into a practical application and regarding Step 2B, there are no additional elements recited that amount to significantly more than the judicial exception.
Regarding Claim 12
Step 2A, Prong 2
Regarding the “wherein the at least one machine learning artificial intelligence model is chosen from the group consisting of decision tree algorithms, random forest algorithms, support vector machine algorithms, adaptive boosting algorithms, back-propagation algorithms, gradient boosting algorithms, neural network algorithms, and deep neural network algorithms” limitation, such limitation amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use. As explained by the Supreme Court, a claim directed to a judicial exception cannot be made eligible "simply by having the applicant acquiesce to limiting the reach of the patent for the formula to a particular technological use." Diamond v. Diehr, 450 U.S. 175, 192 n.14, 209 USPQ 1, 10 n. 14 (1981). Thus, limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception do not integrate a judicial exception into a practical application.
Step 2B
Regarding the “wherein the at least one machine learning artificial intelligence model is chosen from the group consisting of decision tree algorithms, random forest algorithms, support vector machine algorithms, adaptive boosting algorithms, back-propagation algorithms, gradient boosting algorithms, neural network algorithms, and deep neural network algorithms” limitation, such limitation amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use as explained above, which does not amount to significantly more than the judicial exception. MPEP 2106.05(h).
Regarding Claim 13
Step 2A, prong 1 (Is the claim directed to a law of nature, a natural phenomenon or an abstract idea).
Claim 13 recites the following mental processes, that in each case under the broadest reasonable interpretation, covers performance of the limitation in the mind (including an observation, evaluation, judgment, opinion) or with the aid of pencil and paper but for the recitation of generic computer components (e.g., “machine learning artificial intelligence model”, “memory”, “consistency controller”, “weight calculation controller”).
carry out at least one predetermined consistency test to check that a reliable reference weight has been calculated or is capable of being calculated; (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by comparing a measured weight to a reliable reference weight to determine if the measured weight is capable of being reliably calculated)
calculate at least one calculated weight of the aircraft as a function of at least one weight chosen from the group consisting of a current boarded fuel weight, a consumed fuel weight, a previously measured weight of the aircraft, an empty weight of the aircraft, a payload previously input by a user, a weight measured by a piece of connected equipment, and the reference weight; (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by using a mathematical calculation, using the recited parameters as dependent variables, to calculate a predicted weight of the aircraft)
exclude sets of flight data that do not satisfy the consistency test from the training data so as to improve reliability of the training data; and use the sets of training data to program the at least one machine learning artificial intelligence model, (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, e.g., by excluding unreliable flight data from being used to train a neural network model, thus improving the output of estimation of aircraft weight during operation by selecting only reliable flight data)
Step 2A, prong 2 (Does the claim recite additional elements that integrate the judicial exception into a practical application?).
The judicial exception is not integrated into a practical application. In particular, the claim recites the additional elements (e.g., “machine learning artificial intelligence model”, “memory”, “consistency controller”, “weight calculation controller”) which are recited at a high-level of generality such that they amount to no more than mere instructions to apply the exception using a generic computer component (See MPEP 2106.05(f)).
Regarding the “an aircraft” limitation, such limitation amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use. As explained by the Supreme Court, a claim directed to a judicial exception cannot be made eligible "simply by having the applicant acquiesce to limiting the reach of the patent for the formula to a particular technological use." Diamond v. Diehr, 450 U.S. 175, 192 n.14, 209 USPQ 1, 10 n. 14 (1981). Thus, limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception do not integrate a judicial exception into a practical application.
Regarding the “a memory configured to store at least one machine learning artificial intelligence model configured to be implemented during at least one predetermined flight phase of the aircraft;” limitation, such additional element of a data storage step is recited at a high level of generality and amounts to extra-solution activity of storing data, i.e. post-solution activity of data storage for use in the claimed process (see MPEP 2106.05(g)).
Regarding the “a plurality of sensing devices configured to acquire, during flight, at least one set of flight data at a same point in time;” limitation, such additional element of a data gathering step is recited at a high level of generality and amounts to extra-solution activity of receiving data, i.e. pre-solution activity of gathering data for use in the claimed process (see MPEP 2106.05(g)).
Regarding the “at least one consistency controller; and at least one weight calculation controller, wherein the at least one consistency controller and the at least one weight calculation controller are configured to:” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of a controller. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (a controller). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Regarding the “store, when the at least one consistency test is validated, the at least one set of flight data and the at least one calculated weight at the same point in time, each stored set of flight data associated with a calculated weight forming a set of training data for the at least one machine learning artificial intelligence model;” limitation, such additional element of a data storage step is recited at a high level of generality and amounts to extra-solution activity of storing data, i.e. post-solution activity of data storage for use in the claimed process (see MPEP 2106.05(g)).
Regarding the “wherein the at least one machine learning artificial intelligence model is configured to estimate, at any time, an estimated instantaneous weight of the aircraft based on a current set of flight data” limitation, such limitation amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (predicting the weight of an aircraft). As explained by the Supreme Court, a claim directed to a judicial exception cannot be made eligible "simply by having the applicant acquiesce to limiting the reach of the patent for the formula to a particular technological use." Diamond v. Diehr, 450 U.S. 175, 192 n.14, 209 USPQ 1, 10 n. 14 (1981). Thus, limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception do not integrate a judicial exception into a practical application.
Step 2B (Does the claim recite additional elements that amount to significantly more than the judicial exception?)
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception. As discussed above, the additional elements (e.g., “machine learning artificial intelligence model”, “memory”, “consistency controller”, “weight calculation controller”)re recited at a high-level of generality such that they amount to no more than mere instructions to apply the exception using a generic computer component (See MPEP 2106.05(f)).
Regarding the “an aircraft” limitation, such limitation amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use as explained above, which does not amount to significantly more than the judicial exception. MPEP 2106.05(h).
Regarding the “a memory configured to store at least one machine learning artificial intelligence model configured to be implemented during at least one predetermined flight phase of the aircraft;” limitation, as discussed above, the additional element of a data storage step is recited at a high level of generality and amounts to extra-solution activity of storing data, i.e. post-solution activity of storing data after or during use in the claimed process. The courts have found limitations directed to storing information electronically, recited at a high level of generality, to be well-understood, routine, and conventional (see MPEP 2106.05(d)(II), "electronic record keeping," and "storing and retrieving information in memory").
Regarding the “a plurality of sensing devices configured to acquire, during flight, at least one set of flight data at a same point in time;” limitation, as discussed above, the additional element of a data gathering step is recited at a high level of generality and amounts to extra-solution activity of receiving data, i.e. pre-solution activity of gathering data for use in the claimed process. The courts have found limitations directed to obtaining information electronically, recited at a high level of generality, to be well-understood, routine, and conventional (see MPEP 2106.05(d)(II), “receiving or transmitting data over a network”, "electronic record keeping," and "storing and retrieving information in memory").
Regarding the “at least one consistency controller; and at least one weight calculation controller, wherein the at least one consistency controller and the at least one weight calculation controller are configured to:” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Regarding the “store, when the at least one consistency test is validated, the at least one set of flight data and the at least one calculated weight at the same point in time, each stored set of flight data associated with a calculated weight forming a set of training data for the at least one machine learning artificial intelligence model;” limitation, as discussed above, the additional element of a data storage step is recited at a high level of generality and amounts to extra-solution activity of storing data, i.e. post-solution activity of storing data after or during use in the claimed process. The courts have found limitations directed to storing information electronically, recited at a high level of generality, to be well-understood, routine, and conventional (see MPEP 2106.05(d)(II), "electronic record keeping," and "storing and retrieving information in memory").
Regarding the “wherein the at least one machine learning artificial intelligence model is configured to estimate, at any time, an estimated instantaneous weight of the aircraft based on a current set of flight data” limitation, such limitation amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use as explained above, which does not amount to significantly more than the judicial exception. MPEP 2106.05(h).
Regarding Claim 14
Step 2A, Prong 1
wherein the reference weight is defined as a function of an estimated weight of the aircraft based on measurements from several sensors, each sensor being arranged on each landing gear of the aircraft, (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by defining a reference weight so that it is determined based on a function of measurement inputs as recited)
calculate a difference between the calculated weight and the estimated weight and to verify that the difference is less than a predetermined threshold value. (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by calculating a difference between the calculated and estimated weight and then comparing that difference to a threshold)
Step 2A, Prong 2
Regarding the “wherein the at least one consistency controller is configured to” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of a controller. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (a controller). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Step 2B
Regarding the “wherein the at least one consistency controller is configured to” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Regarding Claim 15
Step 2A, Prong 1
wherein the reference weight is defined as a function of a checked theoretical weight of the aircraft, (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by defining a reference weight so that it is determined based on a function of a checked theoretical weight)
verify parameterization of the checked theoretical weight. (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by checking the different parameters used to verify the checked theoretical weight)
Step 2A, Prong 2
Regarding the “wherein the at least one consistency controller is configured to” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of a controller. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (a controller). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Step 2B
Regarding the “wherein the at least one consistency controller is configured to” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Regarding Claim 16
Step 2A, Prong 1
wherein the reference weight is defined as a function of a calculated theoretical weight of the aircraft, (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by defining a reference weight so that it is determined based on a function of a calculated theoretical weight)
... verify that the at least one piece of information has been parameterized. (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by checking the different parameters used to verify the calculated theoretical weight)
Step 2A, Prong 2
Regarding the “the calculated theoretical weight being obtained by the at least one weight calculation controller based on at least one piece of information parameterized with a human-machine interface of the aircraft” limitation, such additional element of a data gathering step is recited at a high level of generality and amounts to extra-solution activity of receiving data, i.e. pre-solution activity of gathering data for use in the claimed process (see MPEP 2106.05(g)).
Regarding the “wherein the at least one consistency controller is configured to” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of a controller. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (a controller). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Step 2B
Regarding the “the calculated theoretical weight being obtained by the at least one weight calculation controller based on at least one piece of information parameterized with a human-machine interface of the aircraft” limitation, as discussed above, the additional element of a data gathering step is recited at a high level of generality and amounts to extra-solution activity of receiving data, i.e. pre-solution activity of gathering data for use in the claimed process. The courts have found limitations directed to obtaining information electronically, recited at a high level of generality, to be well-understood, routine, and conventional (see MPEP 2106.05(d)(II), “receiving or transmitting data over a network”, "electronic record keeping," and "storing and retrieving information in memory").
Regarding the “wherein the at least one consistency controller is configured to” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Regarding Claim 17
Step 2A, Prong 2
Regarding the “wherein the flight data has data relating to at least two flight parameters chosen from the group consisting of a speed of the aircraft in relation to the air, a vertical speed of the aircraft in relation to the ground, a longitudinal speed of the aircraft in relation to the ground, a lateral speed of the aircraft in relation to the ground, a vertical acceleration of the aircraft in relation to the ground, a flow of fuel supplying an engine of the aircraft, a rotational speed of a rotor equipping the aircraft, a wind direction, a wind speed, a quantity of fuel on board the aircraft, a yaw trajectory of the aircraft, an attitude of the aircraft, an air density, an air temperature, an altitude of the aircraft, an angle of attack of a wing of the aircraft, a power consumed by at least one engine of the aircraft, positions of flight controls and positions of blades of a rotor and/or of a propeller” limitation, this limitation merely describes the types of data used to estimate aircraft weight, and therefore such limitation amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use. As explained by the Supreme Court, a claim directed to a judicial exception cannot be made eligible "simply by having the applicant acquiesce to limiting the reach of the patent for the formula to a particular technological use." Diamond v. Diehr, 450 U.S. 175, 192 n.14, 209 USPQ 1, 10 n. 14 (1981). Thus, limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception do not integrate a judicial exception into a practical application.
Step 2B
Regarding the “wherein the flight data has data relating to at least two flight parameters chosen from the group consisting of a speed of the aircraft in relation to the air, a vertical speed of the aircraft in relation to the ground, a longitudinal speed of the aircraft in relation to the ground, a lateral speed of the aircraft in relation to the ground, a vertical acceleration of the aircraft in relation to the ground, a flow of fuel supplying an engine of the aircraft, a rotational speed of a rotor equipping the aircraft, a wind direction, a wind speed, a quantity of fuel on board the aircraft, a yaw trajectory of the aircraft, an attitude of the aircraft, an air density, an air temperature, an altitude of the aircraft, an angle of attack of a wing of the aircraft, a power consumed by at least one engine of the aircraft, positions of flight controls, and positions of blades of a rotor and/or of a propeller” limitation, such limitation amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use as explained above, which does not amount to significantly more than the judicial exception. MPEP 2106.05(h).
Regarding Claim 18
Step 2A, Prong 1
the at least one predetermined flight phase is chosen as a function of a required accuracy of the at least one machine learning artificial intelligence model for estimating, at any time, the estimated instantaneous weight of the aircraft based on a current set of flight data. (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by selecting a flight phase to prioritize accuracy of the machine learning model)
Step 2A, Prong 2
Regarding the “wherein the at least one machine learning artificial intelligence model is configured to be implemented during the at least one predetermined flight phase” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of using a machine learning model. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (machine learning models). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Regarding the “wherein the at least one weight calculation controller is configured to apply the at least one machine learning artificial intelligence model during the chosen predetermined flight phase to estimate the estimated instantaneous weight of the aircraft based on the current set of flight data” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of using a machine learning model to estimate an output based on certain input. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (machine learning models). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Step 2B
Regarding the “wherein the at least one machine learning artificial intelligence model is configured to be implemented during the at least one predetermined flight phase” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Regarding the “wherein the at least one weight calculation controller is configured to apply the at least one machine learning artificial intelligence model during the chosen predetermined flight phase to estimate the estimated instantaneous weight of the aircraft based on the current set of flight data” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Regarding Claim 19
Step 2A, Prong 1
the at least one predetermined flight phase being selected as a function of a required dispersion of the at least one machine learning artificial intelligence model for estimating, at any time, the estimated instantaneous weight of the aircraft based on a current set of flight data. (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by selecting a flight phase out of a distribution (or dispersion) of different flight phases, to pick the closest flight phase for the most accurate measurements)
Step 2A, Prong 2
Regarding the “wherein the at least one machine learning artificial intelligence model is configured to be implemented during the at least one predetermined flight phase” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of using a machine learning model. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (machine learning models). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Regarding the “wherein the at least one weight calculation controller is configured to apply the at least one machine learning artificial intelligence model during the chosen predetermined flight phase to estimate the estimated instantaneous weight of the aircraft based on the current set of flight data” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of using a machine learning model to estimate an output based on certain input. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (machine learning models). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Step 2B
Regarding the “wherein the at least one machine learning artificial intelligence model is configured to be implemented during the at least one predetermined flight phase” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Regarding the “wherein the at least one weight calculation controller is configured to apply the at least one machine learning artificial intelligence model during the chosen predetermined flight phase to estimate the estimated instantaneous weight of the aircraft based on the current set of flight data” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Regarding Claim 20
Step 2A, Prong 1
the at least one predetermined flight phase being selected as a function of a diversity of the plurality of flights performed by a user of the aircraft. (under the broadest reasonable interpretation, a human such as a flight engineer can perform this step mentally, such as by selecting a flight phase out of a distribution (or diversity) of different flight phases, to pick the closest flight phase for the most accurate measurements)
Step 2A, Prong 2
Regarding the “wherein the at least one machine learning artificial intelligence model is configured to be implemented during the at least one predetermined flight phase” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception. In particular, the claim only recites the additional element of using a machine learning model. This additional element is recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (machine learning models). Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea (See MPEP 2106.05(f)).
Step 2B
Regarding the “wherein the at least one machine learning artificial intelligence model is configured to be implemented during the at least one predetermined flight phase” limitation, such limitation is recited at a high-level of generality and amounts to no more than adding the words “apply it” (or an equivalent) with the judicial exception, because the limitation merely provides instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, this additional element does not add significantly more than the judicial exception. (See MPEP 2106.05(f)).
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 5, 12, 13, and 17 are rejected under 35 U.S.C. 103 as being unpatentable over US 5987397 A, hereinafter referenced as MCCOOL, in view of US 7624080 B1, hereinafter referenced as MORALES, and further in view of US 20220076404 A1, hereinafter referenced as IMOTO.
Regarding Claim 1
MCCOOL teaches:
A method for training at least one machine learning artificial intelligence model, (MCCOOL, col. 3, lines 43-46: “Training the neural network results in one or more neural network equations being learned for performance of the parameter signal calculations.”) the at least one machine learning artificial intelligence model being configured to be stored in a memory of an aircraft or a ground station, and configured to be implemented during at least one predetermined flight phase of at least one aircraft of the same type, (MCCOOL, col. 1, lines 46-51: “Memory means is provided for storing the learned relationship as a nonlinear algorithm on board the helicopter for use in a signal processor, receiving real time values of the input parameters and in accordance with said algorithm determine and display estimates of the gross weight and center of gravity locations under flight conditions.”;
MCCOOL, col. 2, lines 54-66: “Such relationship is evaluated from a set of test data such as predicted flight data under hover flight conditions at a variety of gross weights at data points distinct from the data points for the aforementioned set of training data in order to assess how well the network system generalizes. ... The system 10 thus identifies through indicator 20 gross weight and center of gravity location during initial takeoff under hover conditions”;
Examiner’s Note: the “hover conditions” of the helicopter corresponds to the recited “predetermined flight phase of the at least one aircraft of the same type”)
the method comprising carrying out a plurality of flights, and, for at least one of the plurality of flights, acquiring, during flight, at least one set of flight data acquired with several sensing devices at the same point in time t, (MCCOOL, col. 2, lines 21-28: “With continued reference to FIG. 1, the input parameter signals in terms of real time values are fed from the inflight measurement means 14 to memory 18 for storage. The signal processor 16 also receives parameter inputs from the measurement means 14 and processes the inputs thereto in accordance with a stored nonlinear mathematical algorithm received from the memory 18 of the input parameter determining means 12.”;
MCCOOL, col. 3, lines 16-20: “The training exemplars, which include the input parameters and a corresponding desired output are either directly measured during test flights or are determined based on parameters measured during test flights.”
MCCOOL, col. 3, lines 28-33: “ At 38, onboard sensors of measurement means 14 measure the variable parameters in the helicopter fixed reference frame. At 40, the gross weight and center of gravity location are calculated within the signal processor 16, based on the measured variable parameters at 38.”
Examiner’s Note: MCCOOL teaches a plurality of test flights, where during such test flights, sensors (plural) (corresponding to recited “several sensing devices”) are used to collect, in real-time (corresponding to recited “at the same point in time t”) measurement data that is used to predict the gross weight and center of gravity of the helicopter)
wherein, for the at least one of the plurality of flights, the method comprises the following steps: (MCCOOL, col. 3, lines 16-20: “The training exemplars, which include the input parameters and a corresponding desired output are either directly measured during test flights or are determined based on parameters measured during test flights.”)
calculating, with at least one weight calculation controller, at least one calculated weight of the aircraft as a function of at least one weight chosen from the group consisting of a current boarded fuel weight, a consumed fuel weight, a previously measured weight of the aircraft, an empty weight of the aircraft, a payload previously input by a user, a weight measured by a piece of connected equipment, and the reference weight; and (MCCOOL, col. 2, lines 32-40: “The variable parameter inputs from measuring means 14 are based on a plurality of measurements potentially including but not limited to: (1) collective stick position; (2) lateral cyclic stick position; (3) longitudinal cyclic stick position; (4) pedal position; (5) pitch attitude; (6) roll attitude; (7) pitch rate; (8) roll rate; (9) yaw rate; (10) engine torque; (11) rotor speed; (12) altitude; (13) load factor; (14) rate of climb; (15) fuel flow; and (16) fuel weight. Other of such input data measurements may also be required.”;
MCCOOL, col. 3, lines 28-33: “ At 38, onboard sensors of measurement means 14 measure the variable parameters in the helicopter fixed reference frame. At 40, the gross weight and center of gravity location are calculated within the signal processor 16, based on the measured variable parameters at 38.”
Examiner’s Note: MCCOOL teaches a signal processor (corresponding to recited “weight calculation controller”) that calculates the gross weight of a helicopter as a function of the fuel weight (corresponding to recited “boarded fuel weight”) and fuel flow (which can be used to determine the recited “consumed fuel weight”))
using the sets of training data to program the at least one machine learning artificial intelligence model, (MCCOOL, col. 2, lines 49-58: “Based on such training data, a sufficient level of correlation is achieved between predicted and measured gross weight to establish a mathematical algorithm relationship between the variable parameter inputs and aircraft gross weight. Such relationship is evaluated from a set of test data such as predicted flight data under hover flight conditions at a variety of gross weights at data points distinct from the data points for the aforementioned set of training data in order to assess how well the network system generalizes.”;
MCCOOL, col. 3, lines 14-25: “ Next at 32, training exemplars corresponding to gross weight and location of center of gravity used to train the network are determined. The training exemplars, which include the input parameters and a corresponding desired output are either directly measured during test flights or are determined based on parameters measured during test flights. The data used to determine the training exemplars is measured under a plurality of flight conditions. Then, at 34, the neural network learns an input-output relationship between the input parameters and the corresponding desired output such as gross weight and center of gravity location, represented by at least one nonlinear equation.”)
the at least one machine learning artificial intelligence model being configured to estimate, at any time, an estimated instantaneous weight of the aircraft, or another aircraft of the same type, based on a current set of flight data. (MCCOOL, col. 3, lines 28-33: “ At 38, onboard sensors of measurement means 14 measure the variable parameters in the helicopter fixed reference frame. At 40, the gross weight and center of gravity location are calculated within the signal processor 16, based on the measured variable parameters at 38.”
However, MCCOOL fails to explicitly teach:
the method improving reliability of flight data used for training and thereby improving estimation of aircraft weight during operation
carrying out at least one predetermined consistency test with at least one consistency controller, the at least one consistency test configured to check that a reliable reference weight has been calculated or is capable of being calculated;
if the at least one consistency test is validated, storing the at least one set of flight data and the at least one calculated weight at the point in time t, each stored set of flight data associated with a calculated weight forming a set of training data for the at least one machine learning artificial intelligence model, and
wherein sets of flight data that do not satisfy the consistency test are excluded from the training data so as to improve reliability of the training data used by the machine learning artificial intelligence model
However, in a related field of endeavor (adaptive sensors for determining aircraft gross weight, see col. 15, lines 20-25), MORALES teaches and makes obvious:
carrying out at least one predetermined consistency test with at least one consistency controller, the at least one consistency test configured to check that a reliable reference weight has been calculated or is capable of being calculated; (MORALES, col. 8, lines 18-35: “Algorithm 3000 performs a first threshold determination, viz., as to whether a newly rendered (current) measurement 2000 by a sensing element 200 is "known" (synonymously referred to herein as "recognized," "recognizable" or "familiar") or "unkown" (synonymously referred to herein as "unrecognized," "unrecognizable" or "unfamiliar"). This first threshold determination includes comparison of the newly rendered measurement 2000 with at least one previously rendered measurement that has been previously assimilated into knowledge base 4000. If the first threshold determination is that the newly rendered measurement 2000 is known, the newly rendered measurement 2000 is compared to a second threshold to determine whether the measurement-associated outcome is consistent with (e.g., within an assigned tolerance of) the previous measurement associated outcome(s). If this second threshold determination is that such consistency exists, the measurement characteristics are assimilated into knowledge base 4000 (which is contained in memory 302).”;
MORALES, col. 15, line 64 – col. 16, line 4: “In accordance with the U.S. Navy's testing of the present invention's aircraft gross weight adaptive sensor, the gross weight information is shown to the pilot through the onboard HUMS cockpit display as it becomes available. A time-stamp defines when the estimation is obtained. During the flight test, the logging of aircraft gross weight is carefully controlled, providing the information necessary for improving the relationship.”
Examiner’s Note: MORALES teaches that the aircraft gross weight is “carefully controlled” during flight tests, which determines the recited “reference weight” and that new measurements by a sensing element (such as the aircraft gross weight adaptive sensor) are compared to a threshold to determine if the measurement is consistent with previous measurements (corresponding to recited “predetermined consistency test with at least one consistency controller”); the MCCOOL-MORALES combination now modifies the neural network that predicts helicopter gross weight of MCCOOL with the teachings of MORALES with respect to collecting test data and only assimilating such testing data into a knowledge base if the measurement is within an accepter tolerance of previously measured results)
if the at least one consistency test is validated, storing the at least one set of flight data and the at least one calculated weight at the point in time t, each stored set of flight data associated with a calculated weight forming a set of training data for the at least one machine learning artificial intelligence model, and (MORALES, col. 8, lines 18-35: “Algorithm 3000 performs a first threshold determination, viz., as to whether a newly rendered (current) measurement 2000 by a sensing element 200 is "known" (synonymously referred to herein as "recognized," "recognizable" or "familiar") or "unkown" (synonymously referred to herein as "unrecognized," "unrecognizable" or "unfamiliar"). This first threshold determination includes comparison of the newly rendered measurement 2000 with at least one previously rendered measurement that has been previously assimilated into knowledge base 4000. If the first threshold determination is that the newly rendered measurement 2000 is known, the newly rendered measurement 2000 is compared to a second threshold to determine whether the measurement-associated outcome is consistent with (e.g., within an assigned tolerance of) the previous measurement associated outcome(s). If this second threshold determination is that such consistency exists, the measurement characteristics are assimilated into knowledge base 4000 (which is contained in memory 302).”;
MCCOOL, col. 12, lines 33-42: “Once an association of newly rendered measurement(s) 2000 has been successfully corroborated, that association is assimilated into knowledge base 4000. The association thereby becomes part of the permanent knowledge base of smart sensing device 100. With the assimilation of this new knowledge, the inventive adaptation process is complete as to this new knowledge, smart sensor 100 thus having successfully improved its capabilities by virtue of enhancing its knowledge base.”;
Examiner’s Note: MORALES teaches that new measurements by a sensing element (such as the aircraft gross weight adaptive sensor) are compared to a threshold to determine if the measurement is consistent with previous measurements, and are only assimilated into a knowledge base of determined to be sufficiently consistent; the MCCOOL-MORALES combination now modifies the neural network that predicts helicopter gross weight of MCCOOL with the teachings of MORALES with respect to collecting test data and only assimilating such testing data into a knowledge base if the measurement is within an accepter tolerance of previously measured results, and then using such knowledge base to train the neural network model of MCCOOL)
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES as explained above. As disclosed by MORALES, one of ordinary skill would have been motivated to do so because MORALES teaches improved “sensor capabilities through effectuation of an adaptive strategy involving the extraction, categorization and fusion of sensor data.” (col. 2, lines 63-66). As further disclosed by MORALES, one of ordinary skill would have been motivated to do so because MORALES teaches that the “knowledge structure utilized by the inventive adaptive sensor is hierarchical and localized and, as such, possesses a natural resiliency against error assimilation.” (col. 3, lines 39-42). One of ordinary skill would understand that the consistency checking performed by MORALES improves the training dataset by only incorporating measurements considered to be reliable.
However, MCCOOL and MORALES fail to explicitly teach:
the method improving reliability of flight data used for training and thereby improving estimation of aircraft weight during operation
wherein sets of flight data that do not satisfy the consistency test are excluded from the training data so as to improve reliability of the training data used by the machine learning artificial intelligence model
However, in a related field of endeavor (machine learning models related to hardware prediction, see paras. 0003-0004), ITOMI teaches and makes obvious:
the method improving reliability of flight data used for training and thereby improving estimation of aircraft weight during operation (ITOMI, para. 0197: “This structure enables removal of unreliable data from the training data according to the probability, and enables achievement of learning using more reliable training data.”;
Examiner’s Note: the MCCOOL-MORALES-ITOMI combination now modifies the neural network that predicts helicopter gross weight of MCCOOL with the teachings of MORALES with respect to collecting test data and only assimilating such testing data into a knowledge base if the measurement is within an acceptable tolerance of previously measured results, and then using such knowledge base to train the neural network model of MCCOOL, because removing unreliable data will enable the achievement of learning using more reliable training data as taught by ITOMI)
wherein sets of flight data that do not satisfy the consistency test are excluded from the training data so as to improve reliability of the training data used by the machine learning artificial intelligence model (ITOMI, para. 0197: “This structure enables removal of unreliable data from the training data according to the probability, and enables achievement of learning using more reliable training data.”;
Examiner’s Note: the MCCOOL-MORALES-ITOMI combination now modifies the neural network that predicts helicopter gross weight of MCCOOL with the teachings of MORALES with respect to collecting test data and only assimilating such testing data into a knowledge base if the measurement is within an acceptable tolerance of previously measured results, and then using such knowledge base to train the neural network model of MCCOOL, because removing unreliable data will enable the achievement of learning using more reliable training data as taught by ITOMI)
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES and ITOMI as explained above. As disclosed by ITOMI, one of ordinary skill would have been motivated to do so in order to generate “a trained model trained using more reliable information.” (para. 0193). One of ordinary skill would further understand that removing unreliable data points from training data (e.g., data where labels are uncertain) will improve the quality of the trained model.
Regarding Claim 5
MCCOOL, MORALES, and ITOMI disclose the method according to claim 1 as explained above. MCCOOL further teaches:
wherein the flight data has data relating to at least two flight parameters chosen from the group consisting of a speed of the aircraft in relation to the air, a vertical speed of the aircraft in relation to the ground, a longitudinal speed of the aircraft in relation to the ground, a lateral speed of the aircraft in relation to the ground, a vertical acceleration of the aircraft in relation to the ground, a flow of fuel supplying an engine of the aircraft, a rotational speed of a rotor equipping the aircraft, a wind direction, a wind speed, a quantity of fuel on board the aircraft, a yaw trajectory of the aircraft, an attitude of the aircraft, an air density, an air temperature, an altitude of the aircraft, an angle of attack of a wing of the aircraft, a power consumed by at least one engine of the aircraft, positions of flight controls and positions of blades of a rotor and/or of a propeller. (MCCOOL, col. 2, lines 32-40: “The variable parameter inputs from measuring means 14 are based on a plurality of measurements potentially including but not limited to: (1) collective stick position; (2) lateral cyclic stick position; (3) longitudinal cyclic stick position; (4) pedal position; (5) pitch attitude; (6) roll attitude; (7) pitch rate; (8) roll rate; (9) yaw rate; (10) engine torque; (11) rotor speed; (12) altitude; (13) load factor; (14) rate of climb; (15) fuel flow; and (16) fuel weight. Other of such input data measurements may also be required.”;
MCCOOL, col. 3, lines 28-33: “ At 38, onboard sensors of measurement means 14 measure the variable parameters in the helicopter fixed reference frame. At 40, the gross weight and center of gravity location are calculated within the signal processor 16, based on the measured variable parameters at 38.”
Examiner’s Note: MCCOOL teaches using the following parameters to predict gross weight during hover conditions: pitch attitude and roll attitude (each corresponding to recited “the attitude of the aircraft”), yaw rate (corresponding to recited “yaw trajectory”), rotor speed (corresponding to recited “rotational speed of a rotor”), altitude (corresponding to recited “an altitude of the aircraft”), fuel flow (corresponding to recited “flow of fuel supplying an engine of the aircraft”) and fuel weight (corresponding to recited “quantity of fuel on board the aircraft”))
Regarding Claim 12
MCCOOL, MORALES, and ITOMI disclose the method according to claim 1 as explained above. MCCOOL further teaches:
wherein the at least one machine learning artificial intelligence model is chosen from the group consisting of decision tree algorithms, random forest algorithms, support vector machine algorithms, adaptive boosting algorithms, back-propagation algorithms, gradient boosting algorithms, neural network algorithms, and deep neural network algorithms. (MCCOOL, col. 3, lines 43-46: “Training the neural network results in one or more neural network equations being learned for performance of the parameter signal calculations.”)
Regarding Claim 13
MCCOOL teaches:
An aircraft comprising: (MCCOOL, col. 1, lines 17-26: “Current health and usage monitoring systems for helicopters require aircraft gross weight and center of gravity location data to accurately calculate estimates on fatigue damage accumulation. Such data is initially entered manually into the monitoring system by the helicopter pilot and/or maintainer. It is therefore an important object of the present invention to eliminate the need for such manual entry of gross weight and center of gravity location data in helicopter health and usage monitoring systems to improve its efficiency as well as to reduce human error.”)
a memory configured to store at least one machine learning artificial intelligence model configured to be implemented during at least one predetermined flight phase of the aircraft; (MCCOOL, col. 1, lines 46-51: “Memory means is provided for storing the learned relationship as a nonlinear algorithm on board the helicopter for use in a signal processor, receiving real time values of the input parameters and in accordance with said algorithm determine and display estimates of the gross weight and center of gravity locations under flight conditions.”;
MCCOOL, col. 2, lines 54-66: “Such relationship is evaluated from a set of test data such as predicted flight data under hover flight conditions at a variety of gross weights at data points distinct from the data points for the aforementioned set of training data in order to assess how well the network system generalizes. ... The system 10 thus identifies through indicator 20 gross weight and center of gravity location during initial takeoff under hover conditions”;
MCCOOL, col. 3, lines 43-46: “Training the neural network results in one or more neural network equations being learned for performance of the parameter signal calculations.”
Examiner’s Note: the trained neural network corresponds to the recited “machine learning artificial intelligence model” and the “hover conditions” of the helicopter corresponds to the recited “predetermined flight phase of the at aircraft”)
a plurality of sensing devices configured to acquire, during flight, at least one set of flight data at a same point in time; (MCCOOL, col. 2, lines 21-28: “With continued reference to FIG. 1, the input parameter signals in terms of real time values are fed from the inflight measurement means 14 to memory 18 for storage. The signal processor 16 also receives parameter inputs from the measurement means 14 and processes the inputs thereto in accordance with a stored nonlinear mathematical algorithm received from the memory 18 of the input parameter determining means 12.”;
MCCOOL, col. 3, lines 16-20: “The training exemplars, which include the input parameters and a corresponding desired output are either directly measured during test flights or are determined based on parameters measured during test flights.”
MCCOOL, col. 3, lines 28-33: “ At 38, onboard sensors of measurement means 14 measure the variable parameters in the helicopter fixed reference frame. At 40, the gross weight and center of gravity location are calculated within the signal processor 16, based on the measured variable parameters at 38.”
Examiner’s Note: MCCOOL teaches a plurality of test flights, where during such test flights, sensors (plural) (corresponding to recited “plurality of sensing devices”) are used to collect, in real-time (corresponding to recited “at a same point in time”) measurement data that is used to predict the gross weight and center of gravity of the helicopter)
at least one weight calculation controller, wherein the ... at least one weight calculation controller are configured to: calculate at least one calculated weight of the aircraft as a function of at least one weight chosen from the group consisting of a current boarded fuel weight, a consumed fuel weight, a previously measured weight of the aircraft, an empty weight of the aircraft, a payload previously input by a user, a weight measured by a piece of connected equipment, and the reference weight; (MCCOOL, col. 2, lines 32-40: “The variable parameter inputs from measuring means 14 are based on a plurality of measurements potentially including but not limited to: (1) collective stick position; (2) lateral cyclic stick position; (3) longitudinal cyclic stick position; (4) pedal position; (5) pitch attitude; (6) roll attitude; (7) pitch rate; (8) roll rate; (9) yaw rate; (10) engine torque; (11) rotor speed; (12) altitude; (13) load factor; (14) rate of climb; (15) fuel flow; and (16) fuel weight. Other of such input data measurements may also be required.”;
MCCOOL, col. 3, lines 28-33: “ At 38, onboard sensors of measurement means 14 measure the variable parameters in the helicopter fixed reference frame. At 40, the gross weight and center of gravity location are calculated within the signal processor 16, based on the measured variable parameters at 38.”
Examiner’s Note: MCCOOL teaches a signal processor (corresponding to recited “weight calculation controller”) that calculates the gross weight of a helicopter as a function of the fuel weight (corresponding to recited “boarded fuel weight”) and fuel flow (which can be used to determine the recited “consumed fuel weight”))
use the sets of training data to program the at least one machine learning artificial intelligence model, (MCCOOL, col. 2, lines 49-58: “Based on such training data, a sufficient level of correlation is achieved between predicted and measured gross weight to establish a mathematical algorithm relationship between the variable parameter inputs and aircraft gross weight. Such relationship is evaluated from a set of test data such as predicted flight data under hover flight conditions at a variety of gross weights at data points distinct from the data points for the aforementioned set of training data in order to assess how well the network system generalizes.”;
MCCOOL, col. 3, lines 14-25: “ Next at 32, training exemplars corresponding to gross weight and location of center of gravity used to train the network are determined. The training exemplars, which include the input parameters and a corresponding desired output are either directly measured during test flights or are determined based on parameters measured during test flights. The data used to determine the training exemplars is measured under a plurality of flight conditions. Then, at 34, the neural network learns an input-output relationship between the input parameters and the corresponding desired output such as gross weight and center of gravity location, represented by at least one nonlinear equation.”)
wherein the at least one machine learning artificial intelligence model is configured to estimate, at any time, an estimated instantaneous weight of the aircraft based on a current set of flight data. (MCCOOL, col. 3, lines 28-33: “ At 38, onboard sensors of measurement means 14 measure the variable parameters in the helicopter fixed reference frame. At 40, the gross weight and center of gravity location are calculated within the signal processor 16, based on the measured variable parameters at 38.”
However, MCCOOL fails to explicitly teach:
at least one consistency controller; and
wherein the at least one consistency controller ... are configured to:
carry out at least one predetermined consistency test to check that a reliable reference weight has been calculated or is capable of being calculated;
store, when the at least one consistency test is validated, the at least one set of flight data and the at least one calculated weight at the same point in time, each stored set of flight data associated with a calculated weight forming a set of training data for the at least one machine learning artificial intelligence model;
exclude sets of flight data that do not satisfy the consistency test from the training data so as to improve reliability of the training data; and
However, in a related field of endeavor (adaptive sensors for determining aircraft gross weight, see col. 15, lines 20-25), MORALES teaches and makes obvious:
at least one consistency controller; and wherein the at least one consistency controller ... are configured to: carry out at least one predetermined consistency test to check that a reliable reference weight has been calculated or is capable of being calculated; (MORALES, col. 8, lines 18-35: “Algorithm 3000 performs a first threshold determination, viz., as to whether a newly rendered (current) measurement 2000 by a sensing element 200 is "known" (synonymously referred to herein as "recognized," "recognizable" or "familiar") or "unkown" (synonymously referred to herein as "unrecognized," "unrecognizable" or "unfamiliar"). This first threshold determination includes comparison of the newly rendered measurement 2000 with at least one previously rendered measurement that has been previously assimilated into knowledge base 4000. If the first threshold determination is that the newly rendered measurement 2000 is known, the newly rendered measurement 2000 is compared to a second threshold to determine whether the measurement-associated outcome is consistent with (e.g., within an assigned tolerance of) the previous measurement associated outcome(s). If this second threshold determination is that such consistency exists, the measurement characteristics are assimilated into knowledge base 4000 (which is contained in memory 302).”;
MORALES, col. 15, line 64 – col. 16, line 4: “In accordance with the U.S. Navy's testing of the present invention's aircraft gross weight adaptive sensor, the gross weight information is shown to the pilot through the onboard HUMS cockpit display as it becomes available. A time-stamp defines when the estimation is obtained. During the flight test, the logging of aircraft gross weight is carefully controlled, providing the information necessary for improving the relationship.”
Examiner’s Note: MORALES teaches that the aircraft gross weight is “carefully controlled” during flight tests, which determines the recited “reference weight” and that new measurements by a sensing element (such as the aircraft gross weight adaptive sensor) are compared to a threshold to determine if the measurement is consistent with previous measurements (corresponding to recited “predetermined consistency test with at least one consistency controller”); the MCCOOL-MORALES combination now modifies the neural network that predicts helicopter gross weight of MCCOOL with the teachings of MORALES with respect to collecting test data and only assimilating such testing data into a knowledge base if the measurement is within an accepter tolerance of previously measured results)
store, when the at least one consistency test is validated, the at least one set of flight data and the at least one calculated weight at the same point in time, each stored set of flight data associated with a calculated weight forming a set of training data for the at least one machine learning artificial intelligence model; (MORALES, col. 8, lines 18-35: “Algorithm 3000 performs a first threshold determination, viz., as to whether a newly rendered (current) measurement 2000 by a sensing element 200 is "known" (synonymously referred to herein as "recognized," "recognizable" or "familiar") or "unkown" (synonymously referred to herein as "unrecognized," "unrecognizable" or "unfamiliar"). This first threshold determination includes comparison of the newly rendered measurement 2000 with at least one previously rendered measurement that has been previously assimilated into knowledge base 4000. If the first threshold determination is that the newly rendered measurement 2000 is known, the newly rendered measurement 2000 is compared to a second threshold to determine whether the measurement-associated outcome is consistent with (e.g., within an assigned tolerance of) the previous measurement associated outcome(s). If this second threshold determination is that such consistency exists, the measurement characteristics are assimilated into knowledge base 4000 (which is contained in memory 302).”;
MCCOOL, col. 12, lines 33-42: “Once an association of newly rendered measurement(s) 2000 has been successfully corroborated, that association is assimilated into knowledge base 4000. The association thereby becomes part of the permanent knowledge base of smart sensing device 100. With the assimilation of this new knowledge, the inventive adaptation process is complete as to this new knowledge, smart sensor 100 thus having successfully improved its capabilities by virtue of enhancing its knowledge base.”;
Examiner’s Note: MORALES teaches that new measurements by a sensing element (such as the aircraft gross weight adaptive sensor) are compared to a threshold to determine if the measurement is consistent with previous measurements, and are only assimilated into a knowledge base of determined to be sufficiently consistent; the MCCOOL-MORALES combination now modifies the neural network that predicts helicopter gross weight of MCCOOL with the teachings of MORALES with respect to collecting test data and only assimilating such testing data into a knowledge base if the measurement is within an accepter tolerance of previously measured results, and then using such knowledge base to train the neural network model of MCCOOL)
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES as explained above. As disclosed by MORALES, one of ordinary skill would have been motivated to do so because MORALES teaches improved “sensor capabilities through effectuation of an adaptive strategy involving the extraction, categorization and fusion of sensor data.” (col. 2, lines 63-66). As further disclosed by MORALES, one of ordinary skill would have been motivated to do so because MORALES teaches that the “knowledge structure utilized by the inventive adaptive sensor is hierarchical and localized and, as such, possesses a natural resiliency against error assimilation.” (col. 3, lines 39-42). One of ordinary skill would understand that the consistency checking performed by MORALES improves the training dataset by only incorporating measurements considered to be reliable.
However, MCCOOL and MORALES fails to explicitly teach:
exclude sets of flight data that do not satisfy the consistency test from the training data so as to improve reliability of the training data;
However, in a related field of endeavor (machine learning models related to hardware prediction, see paras. 0003-0004), ITOMI teaches and makes obvious:
exclude sets of flight data that do not satisfy the consistency test from the training data so as to improve reliability of the training data; (ITOMI, para. 0197: “This structure enables removal of unreliable data from the training data according to the probability, and enables achievement of learning using more reliable training data.”;
Examiner’s Note: the MCCOOL-MORALES-ITOMI combination now modifies the neural network that predicts helicopter gross weight of MCCOOL with the teachings of MORALES with respect to collecting test data and only assimilating such testing data into a knowledge base if the measurement is within an acceptable tolerance of previously measured results, and then using such knowledge base to train the neural network model of MCCOOL, because removing unreliable data will enable the achievement of learning using more reliable training data as taught by ITOMI)
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES and ITOMI as explained above. As disclosed by ITOMI, one of ordinary skill would have been motivated to do so in order to generate “a trained model trained using more reliable information.” (para. 0193). One of ordinary skill would further understand that removing unreliable data points from training data (e.g., data where labels are uncertain) will improve the quality of the trained model.
Claim 17 depends from claim 13 and claims an aircraft that corresponds to the method of claim 5, and is therefore rejected for the same reasons explained above with respect to claims 5 and 13.
Claims 2 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over MCOOL in view of MORALES and ITOMI and further in view of US 20170322069 A1, hereinafter referenced as MASTRIANNI.
Regarding Claim 2
MCCOOL, MORALES, and ITOMI disclose the method according to claim 1 as explained above. However, MCCOOL fails to explicitly teach:
wherein the reference weight is defined as a function of an estimated weight of the aircraft based on measurements from several sensors,
the at least one consistency test comprising calculating a difference between the calculated weight and the estimated weight and then checking that the difference is less than a difference threshold value.
However, in a related field of endeavor (adaptive sensors for determining aircraft gross weight, see col. 15, lines 20-25), MORALES teaches and makes obvious:
wherein the reference weight is defined as a function of an estimated weight of the aircraft based on measurements from several sensors, (MORALES, col. 8, lines 18-35: “Algorithm 3000 performs a first threshold determination, viz., as to whether a newly rendered (current) measurement 2000 by a sensing element 200 is "known" (synonymously referred to herein as "recognized," "recognizable" or "familiar") or "unkown" (synonymously referred to herein as "unrecognized," "unrecognizable" or "unfamiliar"). This first threshold determination includes comparison of the newly rendered measurement 2000 with at least one previously rendered measurement that has been previously assimilated into knowledge base 4000. If the first threshold determination is that the newly rendered measurement 2000 is known, the newly rendered measurement 2000 is compared to a second threshold to determine whether the measurement-associated outcome is consistent with (e.g., within an assigned tolerance of) the previous measurement associated outcome(s). If this second threshold determination is that such consistency exists, the measurement characteristics are assimilated into knowledge base 4000 (which is contained in memory 302).”;
MORALES, col. 15, line 64 – col. 16, line 4: “In accordance with the U.S. Navy's testing of the present invention's aircraft gross weight adaptive sensor, the gross weight information is shown to the pilot through the onboard HUMS cockpit display as it becomes available. A time-stamp defines when the estimation is obtained. During the flight test, the logging of aircraft gross weight is carefully controlled, providing the information necessary for improving the relationship.”
Examiner’s Note: MORALES teaches that the aircraft gross weight is “carefully controlled” during flight tests, which determines the recited “reference weight” based on measurements from sensing devices; the MCCOOL-ITOMI-MORALES combination now modifies the neural network that predicts helicopter gross weight of MCCOOL with the teachings of MORALES with respect to collecting a reference weight using sensor measurements during carefully controlled flight tests)
the at least one consistency test comprising calculating a difference between the calculated weight and the estimated weight and then checking that the difference is less than a difference threshold value. (MORALES, col. 8, lines 18-35: “Algorithm 3000 performs a first threshold determination, viz., as to whether a newly rendered (current) measurement 2000 by a sensing element 200 is "known" (synonymously referred to herein as "recognized," "recognizable" or "familiar") or "unkown" (synonymously referred to herein as "unrecognized," "unrecognizable" or "unfamiliar"). This first threshold determination includes comparison of the newly rendered measurement 2000 with at least one previously rendered measurement that has been previously assimilated into knowledge base 4000. If the first threshold determination is that the newly rendered measurement 2000 is known, the newly rendered measurement 2000 is compared to a second threshold to determine whether the measurement-associated outcome is consistent with (e.g., within an assigned tolerance of) the previous measurement associated outcome(s). If this second threshold determination is that such consistency exists, the measurement characteristics are assimilated into knowledge base 4000 (which is contained in memory 302).”;
Examiner’s Note: MORALES teaches that new measurements by a sensing element (such as the aircraft gross weight adaptive sensor) are compared to a threshold to determine if the measurement is consistent with previous measurements; the MCCOOL-ITOMI-MORALES combination now modifies the neural network that predicts helicopter gross weight of MCCOOL with the teachings of MORALES with respect to collecting test data and only assimilating such testing data into a knowledge base if the measurement is within an accepter tolerance of previously measured results, where such tolerance is determined by comparing to a tolerance threshold)
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES and ITOMI as explained above. As disclosed by MORALES, one of ordinary skill would have been motivated to do so because MORALES teaches improved “sensor capabilities through effectuation of an adaptive strategy involving the extraction, categorization and fusion of sensor data.” (col. 2, lines 63-66). As further disclosed by MORALES, one of ordinary skill would have been motivated to do so because MOREALES teaches that the “knowledge structure utilized by the inventive adaptive sensor is hierarchical and localized and, as such, possesses a natural resiliency against error assimilation.” (col. 3, lines 39-42). One of ordinary skill would understand that the consistency checking performed by MORALES improves the training dataset by only incorporating measurements considered to be reliable.
However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
each sensor being arranged on each landing gear of the aircraft,
However, in a related field of endeavor (determining gross weight and center of gravity of aircraft, see para. 0003), MASTRIANNI teaches:
each sensor being arranged on each landing gear of the aircraft (MASTRIANNI, para. 0006 “a gross weight sensor is arrangeable on the landing gear assembly”;
Examiner’s Note: the MCCOOL-MORALES-ITOMI-MASTRIANNI combination now attaches the gross weight adaptive sensor of MORALES to the landing gear of the helicopter of MCCOOL as disclosed by MASTRIANNI)
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI and MASTRIANNI as explained above. As disclosed by MASTRIANNI, one of ordinary skill would have been motivated to do so in order to detect a transition from air mode to ground mode in a helicopter. (para. 0002).
Regarding Claim 14
MCCOOL, MORALES, and ITOMI disclose the aircraft according to claim 13 as explained above. However, MCCOOL fails to explicitly teach:
wherein the reference weight is defined as a function of an estimated weight of the aircraft based on measurements from several sensors,
each sensor being arranged on each landing gear of the aircraft
wherein the at least one consistency controller is configured to calculate a difference between the calculated weight and the estimated weight and to verify that the difference is less than a predetermined threshold value.
However, in a related field of endeavor (adaptive sensors for determining aircraft gross weight, see col. 15, lines 20-25), MORALES teaches and makes obvious:
wherein the reference weight is defined as a function of an estimated weight of the aircraft based on measurements from several sensors, (MORALES, col. 8, lines 18-35: “Algorithm 3000 performs a first threshold determination, viz., as to whether a newly rendered (current) measurement 2000 by a sensing element 200 is "known" (synonymously referred to herein as "recognized," "recognizable" or "familiar") or "unkown" (synonymously referred to herein as "unrecognized," "unrecognizable" or "unfamiliar"). This first threshold determination includes comparison of the newly rendered measurement 2000 with at least one previously rendered measurement that has been previously assimilated into knowledge base 4000. If the first threshold determination is that the newly rendered measurement 2000 is known, the newly rendered measurement 2000 is compared to a second threshold to determine whether the measurement-associated outcome is consistent with (e.g., within an assigned tolerance of) the previous measurement associated outcome(s). If this second threshold determination is that such consistency exists, the measurement characteristics are assimilated into knowledge base 4000 (which is contained in memory 302).”;
MORALES, col. 15, line 64 – col. 16, line 4: “In accordance with the U.S. Navy's testing of the present invention's aircraft gross weight adaptive sensor, the gross weight information is shown to the pilot through the onboard HUMS cockpit display as it becomes available. A time-stamp defines when the estimation is obtained. During the flight test, the logging of aircraft gross weight is carefully controlled, providing the information necessary for improving the relationship.”
Examiner’s Note: MORALES teaches that the aircraft gross weight is “carefully controlled” during flight tests, which determines the recited “reference weight” based on measurements from sensing devices; the MCCOOL-ITOMI-MORALES combination now modifies the neural network that predicts helicopter gross weight of MCCOOL with the teachings of MORALES with respect to collecting a reference weight using sensor measurements during carefully controlled flight tests)
wherein the at least one consistency controller is configured to calculate a difference between the calculated weight and the estimated weight and to verify that the difference is less than a predetermined threshold value. (MORALES, col. 8, lines 18-35: “Algorithm 3000 performs a first threshold determination, viz., as to whether a newly rendered (current) measurement 2000 by a sensing element 200 is "known" (synonymously referred to herein as "recognized," "recognizable" or "familiar") or "unkown" (synonymously referred to herein as "unrecognized," "unrecognizable" or "unfamiliar"). This first threshold determination includes comparison of the newly rendered measurement 2000 with at least one previously rendered measurement that has been previously assimilated into knowledge base 4000. If the first threshold determination is that the newly rendered measurement 2000 is known, the newly rendered measurement 2000 is compared to a second threshold to determine whether the measurement-associated outcome is consistent with (e.g., within an assigned tolerance of) the previous measurement associated outcome(s). If this second threshold determination is that such consistency exists, the measurement characteristics are assimilated into knowledge base 4000 (which is contained in memory 302).”;
Examiner’s Note: MORALES teaches that new measurements by a sensing element (such as the aircraft gross weight adaptive sensor) are compared to a threshold to determine if the measurement is consistent with previous measurements; the MCCOOL-ITOMI-MORALES combination now modifies the neural network that predicts helicopter gross weight of MCCOOL with the teachings of MORALES with respect to collecting test data and only assimilating such testing data into a knowledge base if the measurement is within an accepter tolerance of previously measured results, where such tolerance is determined by comparing to a tolerance threshold)
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES and ITOMI as explained above. As disclosed by MORALES, one of ordinary skill would have been motivated to do so because MORALES teaches improved “sensor capabilities through effectuation of an adaptive strategy involving the extraction, categorization and fusion of sensor data.” (col. 2, lines 63-66). As further disclosed by MORALES, one of ordinary skill would have been motivated to do so because MOREALES teaches that the “knowledge structure utilized by the inventive adaptive sensor is hierarchical and localized and, as such, possesses a natural resiliency against error assimilation.” (col. 3, lines 39-42). One of ordinary skill would understand that the consistency checking performed by MORALES improves the training dataset by only incorporating measurements considered to be reliable.
However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
each sensor being arranged on each landing gear of the aircraft
However, in a related field of endeavor (determining gross weight and center of gravity of aircraft, see para. 0003), MASTRIANNI teaches:
each sensor being arranged on each landing gear of the aircraft (MASTRIANNI, para. 0006 “a gross weight sensor is arrangeable on the landing gear assembly”;
Examiner’s Note: the MCCOOL-MORALES-ITOMI-MASTRIANNI combination now attaches the gross weight adaptive sensor of MORALES to the landing gear of the helicopter of MCCOOL as disclosed by MASTRIANNI)
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI and MASTRIANNI as explained above. As disclosed by MASTRIANNI, one of ordinary skill would have been motivated to do so in order to detect a transition from air mode to ground mode in a helicopter. (para. 0002).
Claims 3-4 and 15-16 are rejected under 35 U.S.C. 103 as being unpatentable over MCCOOL in view of MORALES and ITOMI and further in view of US 5548517 A, hereinafter referenced as NANCE.
Regarding Claim 3
MCCOOL, MORALES, and ITOMI disclose the method according to claim 1 as explained above. However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
wherein the reference weight is defined as a function of a checked theoretical weight of the aircraft, the at least one consistency test comprising checking the parameterization of the checked theoretical weight.
However, in a related field of endeavor (determining aircraft weight, see col. 1, lines 33-34), NANCE teaches and makes obvious:
wherein the reference weight is defined as a function of a checked theoretical weight of the aircraft, the at least one consistency test comprising checking the parameterization of the checked theoretical weight. (NANCE, col. 9, lines 4-10: “Subsequent weight determinations by this new invention will use the airplane's empty weight as a starting point, with additional loaded weight calculated and then added to this starting weight to generate a total airplane weight. For example, the pressure transducers 49 are calibrated to measure only that pressure relating to weight in excess of the aircraft's empty weight.”;
Examiner’s Note: NANCE teaches knowing an aircraft’s empty weight (corresponding to recited “theoretical weight”) and then calculating a total weight based on said weight (e.g., adding weight of fuel, payload, passengers, etc.); the MCCOOL-MORALES-ITOMI-NANCE combination now modifies the known good weight of MORALES to be a weight calculated from the empty aircraft weight as in NANCE, where MORALES teaches consistency tests that now also check the consistency of the individual parameters added to the empty weight (e.g., weight of fuel, payload, passengers))
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI, and NANCE as explained above. As disclosed by NANCE, one of ordinary skill would have been motivated to do so in order to provide more accurate weight readings to pilots to give such pilots better data about whether or not to attempt a take-off. NANCE specifically says “The decision whether or not to attempt a take-off, ultimately is made by the pilot in command. This new system will give more accurate information, which can be used to make that decision.” (col. 1, line 66 – col. 2, line 3).
Regarding Claim 4
MCCOOL, MORALES, and ITOMI disclose the method according to claim 1 as explained above. However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
wherein the reference weight is defined as a function of a calculated theoretical weight of the aircraft, the calculated theoretical weight being obtained by the at least one weight calculation controller based on at least one piece of information parameterized with a human-machine interface of the aircraft by the user prior to the at least one consistency test, the at least one consistency test comprising checking that the at least one piece of information has been parameterized.
However, in a related field of endeavor (determining aircraft weight, see col. 1, lines 33-34), NANCE teaches and makes obvious:
wherein the reference weight is defined as a function of a calculated theoretical weight of the aircraft, the calculated theoretical weight being obtained by the at least one weight calculation controller based on at least one piece of information parameterized with a human-machine interface of the aircraft by the user prior to the at least one consistency test, the at least one consistency test comprising checking that the at least one piece of information has been parameterized. (NANCE, col. 9, lines 4-10: “Subsequent weight determinations by this new invention will use the airplane's empty weight as a starting point, with additional loaded weight calculated and then added to this starting weight to generate a total airplane weight. For example, the pressure transducers 49 are calibrated to measure only that pressure relating to weight in excess of the aircraft's empty weight.”;
NANCE, col. 10, lines 6-7: “The results of the calculations for weight, % MAC and confidence are transmitted to the cockpit display 29 (FIG. 3).”
NANCE, col. 10, lines 39-48: “Once the aircraft loading has been completed and all deicing procedures have been implemented, the pilot can then save within this program, the aircraft's current "clean loaded weight". If take-off delays force the aircraft to wait and allow the re-accumulation of ice deposits on exterior surface areas, those accumulations can be indicated in real time as they relate to added weight shown on this system. The pilot may recall the "clean loaded weight" and compare it to existing weight at any time prior to take-off.”
Examiner’s Note: NANCE teaches calculating a theoretical aircraft weight, starting with an empty aircraft weight and adding additional loaded weight (e.g., weight of fuel, payload, ice on wings, etc.), where such information can be displayed on a cockpit interface (corresponding to recited “human-machine interface of the aircraft), where MORALES teaches consistency tests that now also check the consistency of the individual parameters added to the empty weight (e.g., weight of fuel, payload, passengers))
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI, and NANCE as explained above. As disclosed by NANCE, one of ordinary skill would have been motivated to do so in order to provide more accurate weight readings to pilots to give such pilots better data about whether or not to attempt a take-off. NANCE specifically says “The decision whether or not to attempt a take-off, ultimately is made by the pilot in command. This new system will give more accurate information, which can be used to make that decision.” (col. 1, line 66 – col. 2, line 3).
Regarding Claim 15
MCCOOL, MORALES, and ITOMI disclose the aircraft according to claim 13 as explained above. However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
wherein the reference weight is defined as a function of a checked theoretical weight of the aircraft, and wherein the at least one consistency controller is configured to verify parameterization of the checked theoretical weight.
However, in a related field of endeavor (determining aircraft weight, see col. 1, lines 33-34), NANCE teaches and makes obvious:
wherein the reference weight is defined as a function of a checked theoretical weight of the aircraft, and wherein the at least one consistency controller is configured to verify parameterization of the checked theoretical weight. (NANCE, col. 9, lines 4-10: “Subsequent weight determinations by this new invention will use the airplane's empty weight as a starting point, with additional loaded weight calculated and then added to this starting weight to generate a total airplane weight. For example, the pressure transducers 49 are calibrated to measure only that pressure relating to weight in excess of the aircraft's empty weight.”;
Examiner’s Note: NANCE teaches knowing an aircraft’s empty weight (corresponding to recited “theoretical weight”) and then calculating a total weight based on said weight (e.g., adding weight of fuel, payload, passengers, etc.); the MCCOOL-MORALES-ITOMI-NANCE combination now modifies the known good weight of MORALES to be a weight calculated from the empty aircraft weight as in NANCE, where MORALES teaches consistency tests that now also check the consistency of the individual parameters added to the empty weight (e.g., weight of fuel, payload, passengers))
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI, and NANCE as explained above. As disclosed by NANCE, one of ordinary skill would have been motivated to do so in order to provide more accurate weight readings to pilots to give such pilots better data about whether or not to attempt a take-off. NANCE specifically says “The decision whether or not to attempt a take-off, ultimately is made by the pilot in command. This new system will give more accurate information, which can be used to make that decision.” (col. 1, line 66 – col. 2, line 3).
Regarding Claim 16
MCCOOL, MORALES, and ITOMI disclose the aircraft according to claim 13 as explained above. However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
wherein the reference weight is defined as a function of a calculated theoretical weight of the aircraft, the calculated theoretical weight being obtained by the at least one weight calculation controller based on at least one piece of information parameterized with a human-machine interface of the aircraft, wherein the at least one consistency controller is configured to verify that the at least one piece of information has been parameterized.
However, in a related field of endeavor (determining aircraft weight, see col. 1, lines 33-34), NANCE teaches and makes obvious:
wherein the reference weight is defined as a function of a calculated theoretical weight of the aircraft, the calculated theoretical weight being obtained by the at least one weight calculation controller based on at least one piece of information parameterized with a human-machine interface of the aircraft, wherein the at least one consistency controller is configured to verify that the at least one piece of information has been parameterized. (NANCE, col. 9, lines 4-10: “Subsequent weight determinations by this new invention will use the airplane's empty weight as a starting point, with additional loaded weight calculated and then added to this starting weight to generate a total airplane weight. For example, the pressure transducers 49 are calibrated to measure only that pressure relating to weight in excess of the aircraft's empty weight.”;
NANCE, col. 10, lines 6-7: “The results of the calculations for weight, % MAC and confidence are transmitted to the cockpit display 29 (FIG. 3).”
NANCE, col. 10, lines 39-48: “Once the aircraft loading has been completed and all deicing procedures have been implemented, the pilot can then save within this program, the aircraft's current "clean loaded weight". If take-off delays force the aircraft to wait and allow the re-accumulation of ice deposits on exterior surface areas, those accumulations can be indicated in real time as they relate to added weight shown on this system. The pilot may recall the "clean loaded weight" and compare it to existing weight at any time prior to take-off.”
Examiner’s Note: NANCE teaches calculating a theoretical aircraft weight, starting with an empty aircraft weight and adding additional loaded weight (e.g., weight of fuel, payload, ice on wings, etc.), where such information can be displayed on a cockpit interface (corresponding to recited “human-machine interface of the aircraft), where MORALES teaches consistency tests that now also check the consistency of the individual parameters added to the empty weight (e.g., weight of fuel, payload, passengers))
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI, and NANCE as explained above. As disclosed by NANCE, one of ordinary skill would have been motivated to do so in order to provide more accurate weight readings to pilots to give such pilots better data about whether or not to attempt a take-off. NANCE specifically says “The decision whether or not to attempt a take-off, ultimately is made by the pilot in command. This new system will give more accurate information, which can be used to make that decision.” (col. 1, line 66 – col. 2, line 3).
Claims 6-7 are rejected under 35 U.S.C. 103 as being unpatentable over MCCOOL in view of MORALES and ITOMI and further in view of US 20150347908 A1, hereinafter referenced as MATHEW.
Regarding Claim 6
MCCOOL, MORALES, and ITOMI disclose the method according to claim 1 as explained above. However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
wherein, prior to the use of the sets of training data, the method comprises a count for counting the number N of the sets of training data and a comparison between the number N and a predetermined threshold value S.
However, in a related field of endeavor (establishing and updating predictive models, see para. 0002), MATHEW teaches and makes obvious:
wherein, prior to the use of the sets of training data, the method comprises a count for counting the number N of the sets of training data and a comparison between the number N and a predetermined threshold value S. (MATHEW, para. 0035: “As shown, the method 500 begins at step 502, where the predictive model API framework 158 identifies that a threshold amount of additional training data has been added to a set of training data on which a predictive model is based. At step 504, the predictive model API framework 158 determines, based on parameters associated with the predictive model, an appropriate time to update the predictive model.”
Examiner’s Note: MATHEW teaches determining that a specific amount of training data exceeds a threshold amount of training data sufficient to update a predictive model; the MCCOOL-MORALES-ITOMI-MATHEW combination now trains the neural network model of MCCOOL when sufficient training data is available as explained by MATHEW)
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI, and MATHEW as explained above. As disclosed by MATHEW, one of ordinary skill would have been motivated to do so because MATHEW teaches conserving computation resources for training by only training and updating models when sufficient training data exists. (para. 0035).
Regarding Claim 7
MCCOOL, MORALES, ITOMI, and MATHEW disclose the method according to claim 7 as explained above. However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
wherein the use of the sets of training data is implemented when the number N is greater than or equal to the predetermined threshold value S.
However, in a related field of endeavor (establishing and updating predictive models, see para. 0002), MATHEW teaches and makes obvious:
wherein the use of the sets of training data is implemented when the number N is greater than or equal to the predetermined threshold value S. (MATHEW, para. 0035: “As shown, the method 500 begins at step 502, where the predictive model API framework 158 identifies that a threshold amount of additional training data has been added to a set of training data on which a predictive model is based. At step 504, the predictive model API framework 158 determines, based on parameters associated with the predictive model, an appropriate time to update the predictive model.”
Examiner’s Note: MATHEW teaches determining that a specific amount of training data exceeds a threshold amount of training data sufficient to update a predictive model; the MCCOOL-MORALES-ITOMI-MATHEW combination now trains the neural network model of MCCOOL when sufficient training data is available as explained by MATHEW)
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI, and MATHEW as explained above. As disclosed by MATHEW, one of ordinary skill would have been motivated to do so because MATHEW teaches conserving computation resources for training by only training and updating models when sufficient training data exists. (para. 0035).
Claims 8-11 and 18-20 are rejected under 35 U.S.C. 103 as being unpatentable over MCCOOL in view of MORALES and ITOMI and further in view of Grabill, Paul, et al. "Helicopter structural life modeling: Flight regime and gross weight estimation." 2007 IEEE Aerospace Conference. IEEE, 2007, hereinafter referenced as GRABILL.
Regarding Claim 8
MCCOOL, MORALES, and ITOMI disclose the method according to claim 1 as explained above. However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
wherein the at least one machine learning artificial intelligence model comprises a first model and a second model different from the first model, the first model being associated with a first predetermined flight phase from the at least one predetermined flight phase and the second model being associated with a second predetermined flight phase from the at least one predetermined flight phase, the first predetermined flight phase being different from the second predetermined flight phase.
However, in a related field of endeavor (helicopter structural life modeling), GRABILL teaches and makes obvious:
wherein the at least one machine learning artificial intelligence model comprises a first model and a second model different from the first model, the first model being associated with a first predetermined flight phase from the at least one predetermined flight phase and the second model being associated with a second predetermined flight phase from the at least one predetermined flight phase, the first predetermined flight phase being different from the second predetermined flight phase. (GRABILL, p. 3, right column: “A hierarchical set of neural networks was devised to simplify training of the neural nets and to improve overall system statistical performance. The individual nets are trained to recognize 6 to 28 regimes. These nets are much smaller and thus easier to train and faster than one net with all 141 regimes. It has been found that in general much better results are usually obtained when a ‘big’ problem can be broken down into ‘smaller’ problems.
The nets are arranged hierarchically in that the Level Flight network has precedence over the other nine networks. As seen in Figure 3, input data is first passed to the Level Flight network for classification. If the input data is not classified as one of the Level Flight classes, the data is passed to the other nine neural networks for classification.”;
Examiner’s Note: GRABILL teaches organizing a neural network for regime recognition into separate sub-neural networks, where the sub-neural networks are organized by flight phases (e.g., level fight, climbing, diving, turning) as shown by Fig. 3; the MCCOOL-MORALES-ITOMI-GRABILL combination now extends the neural network for weight estimation of MCCOOL, which is based on the hover flight phase, to utilize the other flight phases as taught by GRABILL, where for example the climb flight phase is different from the hover flight phase)
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI, and GRABILL as explained above. As disclosed by GRABILL, one of ordinary skill would have been motivated to do so because GRABILL teaches that recognizing the flight regime (e.g., the flight phase) “is an essential part of the implementation of analysis of flight when used with Flight Operations Quality Assurance (FOQA) programs.” One of ordinary skill would further understand that different flight paths would have different types of important flight parameters, and that it would therefore be beneficial to have individual neural networks for each particular type of flight path.
Regarding Claim 9
MCCOOL, MORALES, and ITOMI disclose the method according to claim 1 as explained above. MCCOOL further teaches:
wherein the at least one machine learning artificial intelligence model is applied during the chosen predetermined flight phase to estimate the estimated instantaneous weight based on the current set of flight data (MCCOOL, col. 3, lines 28-33: “ At 38, onboard sensors of measurement means 14 measure the variable parameters in the helicopter fixed reference frame. At 40, the gross weight and center of gravity location are calculated within the signal processor 16, based on the measured variable parameters at 38.”
MCCOOL, col. 3, lines 14-25: “ Next at 32, training exemplars corresponding to gross weight and location of center of gravity used to train the network are determined. The training exemplars, which include the input parameters and a corresponding desired output are either directly measured during test flights or are determined based on parameters measured during test flights. The data used to determine the training exemplars is measured under a plurality of flight conditions. Then, at 34, the neural network learns an input-output relationship between the input parameters and the corresponding desired output such as gross weight and center of gravity location, represented by at least one nonlinear equation.”)
However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
wherein the at least one predetermined flight phase is chosen as a function of a required accuracy of the at least one machine learning artificial intelligence model for estimating, at any time, the estimated instantaneous weight of the aircraft, or an aircraft of the same type, based on a current set of flight data;
However, in a related field of endeavor (helicopter structural life modeling), GRABILL teaches and makes obvious:
wherein the at least one predetermined flight phase is chosen as a function of a required accuracy of the at least one machine learning artificial intelligence model for estimating, at any time, the estimated instantaneous weight of the aircraft, or an aircraft of the same type, based on a current set of flight data. (GRABILL, p. 3, right column: “A hierarchical set of neural networks was devised to simplify training of the neural nets and to improve overall system statistical performance. The individual nets are trained to recognize 6 to 28 regimes. These nets are much smaller and thus easier to train and faster than one net with all 141 regimes. It has been found that in general much better results are usually obtained when a ‘big’ problem can be broken down into ‘smaller’ problems.
The nets are arranged hierarchically in that the Level Flight network has precedence over the other nine networks. As seen in Figure 3, input data is first passed to the Level Flight network for classification. If the input data is not classified as one of the Level Flight classes, the data is passed to the other nine neural networks for classification.”;
Examiner’s Note: GRABILL teaches organizing a neural network for regime recognition into separate sub-neural networks, where the sub-neural networks are organized by flight phases (e.g., level fight, climbing, diving, turning) as shown by Fig. 3, such that a particular sub-neural network is selected so that the most accurate neural network is used; the MCCOOL-MORALES-ITOMI-GRABILL combination now extends the neural network for weight estimation of MCCOOL, which is based on the hover flight phase, to utilize neural networks for other flight phases as taught by GRABILL, where the particular flight phase is identified in order to select the most accurate neural network model as taught by GRABILL.
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI, and GRABILL as explained above. As disclosed by GRABILL, one of ordinary skill would have been motivated to do so because GRABILL teaches that recognizing the flight regime (e.g., the flight phase) “is an essential part of the implementation of analysis of flight when used with Flight Operations Quality Assurance (FOQA) programs.” One of ordinary skill would further understand that different flight paths would have different types of important flight parameters, and that it would therefore be beneficial to have individual neural networks for each particular type of flight path.
Regarding Claim 10
MCCOOL, MORALES, and ITOMI disclose the method according to claim 1 as explained above. MCCOOL further teaches:
wherein the at least one machine learning artificial intelligence model is applied during the chosen predetermined flight phase to estimate the estimated instantaneous weight based on the current set of flight data (MCCOOL, col. 3, lines 28-33: “ At 38, onboard sensors of measurement means 14 measure the variable parameters in the helicopter fixed reference frame. At 40, the gross weight and center of gravity location are calculated within the signal processor 16, based on the measured variable parameters at 38.”
MCCOOL, col. 3, lines 14-25: “ Next at 32, training exemplars corresponding to gross weight and location of center of gravity used to train the network are determined. The training exemplars, which include the input parameters and a corresponding desired output are either directly measured during test flights or are determined based on parameters measured during test flights. The data used to determine the training exemplars is measured under a plurality of flight conditions. Then, at 34, the neural network learns an input-output relationship between the input parameters and the corresponding desired output such as gross weight and center of gravity location, represented by at least one nonlinear equation.”)
However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
wherein the at least one predetermined flight phase is chosen as a function of a required dispersion of the at least one machine learning artificial intelligence model for estimating, at any time, the estimated instantaneous weight of the aircraft, or an aircraft of the same type, based on a current set of flight data.
However, in a related field of endeavor (helicopter structural life modeling), GRABILL teaches and makes obvious:
wherein the at least one predetermined flight phase is chosen as a function of a required dispersion of the at least one machine learning artificial intelligence model for estimating, at any time, the estimated instantaneous weight of the aircraft, or an aircraft of the same type, based on a current set of flight data. (GRABILL, p. 3, right column: “A hierarchical set of neural networks was devised to simplify training of the neural nets and to improve overall system statistical performance. The individual nets are trained to recognize 6 to 28 regimes. These nets are much smaller and thus easier to train and faster than one net with all 141 regimes. It has been found that in general much better results are usually obtained when a ‘big’ problem can be broken down into ‘smaller’ problems.
The nets are arranged hierarchically in that the Level Flight network has precedence over the other nine networks. As seen in Figure 3, input data is first passed to the Level Flight network for classification. If the input data is not classified as one of the Level Flight classes, the data is passed to the other nine neural networks for classification.”;
Examiner’s Note: GRABILL teaches organizing a neural network for regime recognition into separate sub-neural networks, where the sub-neural networks are organized by flight phases (e.g., level fight, climbing, diving, turning) as shown by Fig. 3, such that a particular sub-neural network is selected based on the distribution (or dispersion) of the neural networks into sub-neural networks based on flight phase type; the MCCOOL-MORALES-ITOMI-GRABILL combination now extends the neural network for weight estimation of MCCOOL, which is based on the hover flight phase, to utilize neural networks for other flight phases as taught by GRABILL, where the particular flight phase is identified in order to select the most accurate neural network model as taught by GRABILL.
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI, and GRABILL as explained above. As disclosed by GRABILL, one of ordinary skill would have been motivated to do so because GRABILL teaches that recognizing the flight regime (e.g., the flight phase) “is an essential part of the implementation of analysis of flight when used with Flight Operations Quality Assurance (FOQA) programs.” One of ordinary skill would further understand that different flight paths would have different types of important flight parameters, and that it would therefore be beneficial to have individual neural networks for each particular type of flight path.
Regarding Claim 11
MCCOOL, MORALES, and ITOMI disclose the method according to claim 1 as explained above. However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
wherein the at least one predetermined flight phase is chosen as a function of a diversity of the plurality of flights performed by the user of the aircraft, or an aircraft of the same type.
However, in a related field of endeavor (helicopter structural life modeling), GRABILL teaches and makes obvious:
wherein the at least one predetermined flight phase is chosen as a function of a diversity of the plurality of flights performed by the user of the aircraft, or an aircraft of the same type. (GRABILL, p. 3, right column: “A hierarchical set of neural networks was devised to simplify training of the neural nets and to improve overall system statistical performance. The individual nets are trained to recognize 6 to 28 regimes. These nets are much smaller and thus easier to train and faster than one net with all 141 regimes. It has been found that in general much better results are usually obtained when a ‘big’ problem can be broken down into ‘smaller’ problems.
The nets are arranged hierarchically in that the Level Flight network has precedence over the other nine networks. As seen in Figure 3, input data is first passed to the Level Flight network for classification. If the input data is not classified as one of the Level Flight classes, the data is passed to the other nine neural networks for classification.”;
Examiner’s Note: GRABILL teaches organizing a neural network for regime recognition into separate sub-neural networks, where the sub-neural networks are organized by flight phases (e.g., level fight, climbing, diving, turning) as shown by Fig. 3, such that a particular sub-neural network is selected based on the distribution (or diversity) of flight paths performed; the MCCOOL-MORALES-ITOMI-GRABILL combination now extends the neural network for weight estimation of MCCOOL, which is based on the hover flight phase, to utilize neural networks for other flight phases as taught by GRABILL, where the particular flight phase is identified in order to select the most accurate neural network model as taught by GRABILL.
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI, and GRABILL as explained above. As disclosed by GRABILL, one of ordinary skill would have been motivated to do so because GRABILL teaches that recognizing the flight regime (e.g., the flight phase) “is an essential part of the implementation of analysis of flight when used with Flight Operations Quality Assurance (FOQA) programs.” One of ordinary skill would further understand that different flight paths would have different types of important flight parameters, and that it would therefore be beneficial to have individual neural networks for each particular type of flight path.
Regarding Claim 18
MCCOOL, MORALES, and ITOMI disclose the aircraft according to claim 13 as explained above. MCCOOL further teaches:
wherein the at least one machine learning artificial intelligence model is configured to be implemented during the at least one predetermined flight phase, ... wherein the at least one weight calculation controller is configured to apply the at least one machine learning artificial intelligence model during the chosen predetermined flight phase to estimate the estimated instantaneous weight of the aircraft based on the current set of flight data. (MCCOOL, col. 1, lines 46-51: “Memory means is provided for storing the learned relationship as a nonlinear algorithm on board the helicopter for use in a signal processor, receiving real time values of the input parameters and in accordance with said algorithm determine and display estimates of the gross weight and center of gravity locations under flight conditions.”;
MCCOOL, col. 2, lines 54-66: “Such relationship is evaluated from a set of test data such as predicted flight data under hover flight conditions at a variety of gross weights at data points distinct from the data points for the aforementioned set of training data in order to assess how well the network system generalizes. ... The system 10 thus identifies through indicator 20 gross weight and center of gravity location during initial takeoff under hover conditions”;
MCCOOL, col. 3, lines 28-33: “ At 38, onboard sensors of measurement means 14 measure the variable parameters in the helicopter fixed reference frame. At 40, the gross weight and center of gravity location are calculated within the signal processor 16, based on the measured variable parameters at 38.”
MCCOOL, col. 3, lines 14-25: “ Next at 32, training exemplars corresponding to gross weight and location of center of gravity used to train the network are determined. The training exemplars, which include the input parameters and a corresponding desired output are either directly measured during test flights or are determined based on parameters measured during test flights. The data used to determine the training exemplars is measured under a plurality of flight conditions. Then, at 34, the neural network learns an input-output relationship between the input parameters and the corresponding desired output such as gross weight and center of gravity location, represented by at least one nonlinear equation.”)
However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
the at least one predetermined flight phase is chosen as a function of a required accuracy of the at least one machine learning artificial intelligence model for estimating, at any time, the estimated instantaneous weight of the aircraft based on a current set of flight data;
However, in a related field of endeavor (helicopter structural life modeling), GRABILL teaches and makes obvious:
the at least one predetermined flight phase is chosen as a function of a required accuracy of the at least one machine learning artificial intelligence model for estimating, at any time, the estimated instantaneous weight of the aircraft based on a current set of flight data; (GRABILL, p. 3, right column: “A hierarchical set of neural networks was devised to simplify training of the neural nets and to improve overall system statistical performance. The individual nets are trained to recognize 6 to 28 regimes. These nets are much smaller and thus easier to train and faster than one net with all 141 regimes. It has been found that in general much better results are usually obtained when a ‘big’ problem can be broken down into ‘smaller’ problems.
The nets are arranged hierarchically in that the Level Flight network has precedence over the other nine networks. As seen in Figure 3, input data is first passed to the Level Flight network for classification. If the input data is not classified as one of the Level Flight classes, the data is passed to the other nine neural networks for classification.”;
Examiner’s Note: GRABILL teaches organizing a neural network for regime recognition into separate sub-neural networks, where the sub-neural networks are organized by flight phases (e.g., level fight, climbing, diving, turning) as shown by Fig. 3, such that a particular sub-neural network is selected so that the most accurate neural network is used; the MCCOOL-MORALES-ITOMI-GRABILL combination now extends the neural network for weight estimation of MCCOOL, which is based on the hover flight phase, to utilize neural networks for other flight phases as taught by GRABILL, where the particular flight phase is identified in order to select the most accurate neural network model as taught by GRABILL.
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI, and GRABILL as explained above. As disclosed by GRABILL, one of ordinary skill would have been motivated to do so because GRABILL teaches that recognizing the flight regime (e.g., the flight phase) “is an essential part of the implementation of analysis of flight when used with Flight Operations Quality Assurance (FOQA) programs.” One of ordinary skill would further understand that different flight paths would have different types of important flight parameters, and that it would therefore be beneficial to have individual neural networks for each particular type of flight path.
Regarding Claim 19
MCCOOL, MORALES, and ITOMI disclose the aircraft according to claim 13 as explained above. MCCOOL further teaches:
wherein the at least one machine learning artificial intelligence model is configured to be implemented during the at least one predetermined flight phase, ... wherein the at least one weight calculation controller is configured to apply the at least one machine learning artificial intelligence model during the chosen predetermined flight phase to estimate the estimated instantaneous weight of the aircraft based on the current set of flight data. (MCCOOL, col. 1, lines 46-51: “Memory means is provided for storing the learned relationship as a nonlinear algorithm on board the helicopter for use in a signal processor, receiving real time values of the input parameters and in accordance with said algorithm determine and display estimates of the gross weight and center of gravity locations under flight conditions.”;
MCCOOL, col. 2, lines 54-66: “Such relationship is evaluated from a set of test data such as predicted flight data under hover flight conditions at a variety of gross weights at data points distinct from the data points for the aforementioned set of training data in order to assess how well the network system generalizes. ... The system 10 thus identifies through indicator 20 gross weight and center of gravity location during initial takeoff under hover conditions”;
MCCOOL, col. 3, lines 28-33: “ At 38, onboard sensors of measurement means 14 measure the variable parameters in the helicopter fixed reference frame. At 40, the gross weight and center of gravity location are calculated within the signal processor 16, based on the measured variable parameters at 38.”
MCCOOL, col. 3, lines 14-25: “ Next at 32, training exemplars corresponding to gross weight and location of center of gravity used to train the network are determined. The training exemplars, which include the input parameters and a corresponding desired output are either directly measured during test flights or are determined based on parameters measured during test flights. The data used to determine the training exemplars is measured under a plurality of flight conditions. Then, at 34, the neural network learns an input-output relationship between the input parameters and the corresponding desired output such as gross weight and center of gravity location, represented by at least one nonlinear equation.”)
However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
the at least one predetermined flight phase being selected as a function of a required dispersion of the at least one machine learning artificial intelligence model for estimating, at any time, the estimated instantaneous weight of the aircraft based on a current set of flight data;
However, in a related field of endeavor (helicopter structural life modeling), GRABILL teaches and makes obvious:
the at least one predetermined flight phase being selected as a function of a required dispersion of the at least one machine learning artificial intelligence model for estimating, at any time, the estimated instantaneous weight of the aircraft based on a current set of flight data; (GRABILL, p. 3, right column: “A hierarchical set of neural networks was devised to simplify training of the neural nets and to improve overall system statistical performance. The individual nets are trained to recognize 6 to 28 regimes. These nets are much smaller and thus easier to train and faster than one net with all 141 regimes. It has been found that in general much better results are usually obtained when a ‘big’ problem can be broken down into ‘smaller’ problems.
The nets are arranged hierarchically in that the Level Flight network has precedence over the other nine networks. As seen in Figure 3, input data is first passed to the Level Flight network for classification. If the input data is not classified as one of the Level Flight classes, the data is passed to the other nine neural networks for classification.”;
Examiner’s Note: GRABILL teaches organizing a neural network for regime recognition into separate sub-neural networks, where the sub-neural networks are organized by flight phases (e.g., level fight, climbing, diving, turning) as shown by Fig. 3, such that a particular sub-neural network is selected based on the distribution (or dispersion) of the neural networks into sub-neural networks based on flight phase type; the MCCOOL-MORALES-ITOMI-GRABILL combination now extends the neural network for weight estimation of MCCOOL, which is based on the hover flight phase, to utilize neural networks for other flight phases as taught by GRABILL, where the particular flight phase is identified in order to select the most accurate neural network model as taught by GRABILL.
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI, and GRABILL as explained above. As disclosed by GRABILL, one of ordinary skill would have been motivated to do so because GRABILL teaches that recognizing the flight regime (e.g., the flight phase) “is an essential part of the implementation of analysis of flight when used with Flight Operations Quality Assurance (FOQA) programs.” One of ordinary skill would further understand that different flight paths would have different types of important flight parameters, and that it would therefore be beneficial to have individual neural networks for each particular type of flight path.
Regarding Claim 20
MCCOOL, MORALES, and ITOMI disclose the aircraft according to claim 1 as explained above. MCCOOL further teaches:
wherein the at least one machine learning artificial intelligence model is configured to be implemented during the at least one predetermined flight phase, (MCCOOL, col. 1, lines 46-51: “Memory means is provided for storing the learned relationship as a nonlinear algorithm on board the helicopter for use in a signal processor, receiving real time values of the input parameters and in accordance with said algorithm determine and display estimates of the gross weight and center of gravity locations under flight conditions.”;
MCCOOL, col. 2, lines 54-66: “Such relationship is evaluated from a set of test data such as predicted flight data under hover flight conditions at a variety of gross weights at data points distinct from the data points for the aforementioned set of training data in order to assess how well the network system generalizes. ... The system 10 thus identifies through indicator 20 gross weight and center of gravity location during initial takeoff under hover conditions”;
MCCOOL, col. 3, lines 28-33: “ At 38, onboard sensors of measurement means 14 measure the variable parameters in the helicopter fixed reference frame. At 40, the gross weight and center of gravity location are calculated within the signal processor 16, based on the measured variable parameters at 38.”
MCCOOL, col. 3, lines 14-25: “ Next at 32, training exemplars corresponding to gross weight and location of center of gravity used to train the network are determined. The training exemplars, which include the input parameters and a corresponding desired output are either directly measured during test flights or are determined based on parameters measured during test flights. The data used to determine the training exemplars is measured under a plurality of flight conditions. Then, at 34, the neural network learns an input-output relationship between the input parameters and the corresponding desired output such as gross weight and center of gravity location, represented by at least one nonlinear equation.”)
However, MCCOOL, MORALES, and ITOMI fail to explicitly teach:
the at least one predetermined flight phase being selected as a function of a diversity of the plurality of flights performed by a user of the aircraft.
However, in a related field of endeavor (helicopter structural life modeling), GRABILL teaches and makes obvious:
the at least one predetermined flight phase being selected as a function of a diversity of the plurality of flights performed by a user of the aircraft. (GRABILL, p. 3, right column: “A hierarchical set of neural networks was devised to simplify training of the neural nets and to improve overall system statistical performance. The individual nets are trained to recognize 6 to 28 regimes. These nets are much smaller and thus easier to train and faster than one net with all 141 regimes. It has been found that in general much better results are usually obtained when a ‘big’ problem can be broken down into ‘smaller’ problems.
The nets are arranged hierarchically in that the Level Flight network has precedence over the other nine networks. As seen in Figure 3, input data is first passed to the Level Flight network for classification. If the input data is not classified as one of the Level Flight classes, the data is passed to the other nine neural networks for classification.”;
Examiner’s Note: GRABILL teaches organizing a neural network for regime recognition into separate sub-neural networks, where the sub-neural networks are organized by flight phases (e.g., level fight, climbing, diving, turning) as shown by Fig. 3, such that a particular sub-neural network is selected based on the distribution (or diversity) of flight paths performed; the MCCOOL-MORALES-ITOMI-GRABILL combination now extends the neural network for weight estimation of MCCOOL, which is based on the hover flight phase, to utilize neural networks for other flight phases as taught by GRABILL, where the particular flight phase is identified in order to select the most accurate neural network model as taught by GRABILL.
Before the effective filing date of the present application, it would have been obvious to one of ordinary skill in the art to combine the teachings of MCCOOL with respect to a neural network that predicts helicopter gross weight with the teachings of MORALES, ITOMI, and GRABILL as explained above. As disclosed by GRABILL, one of ordinary skill would have been motivated to do so because GRABILL teaches that recognizing the flight regime (e.g., the flight phase) “is an essential part of the implementation of analysis of flight when used with Flight Operations Quality Assurance (FOQA) programs.” One of ordinary skill would further understand that different flight paths would have different types of important flight parameters, and that it would therefore be beneficial to have individual neural networks for each particular type of flight path.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US 20180000408 A1 (Heinrich). “When such training examples are excluded from training the detection of sleep states, this type of context information increases the fraction of remaining training examples that correspond to actual sleep states, thereby increasing the reliability of the detection.” (para. 0067).
US 20200202232 A1 (Nagahara). “The production record data analyzer excludes data concerning less reliable events from the production record data and creates a new dispatching rule model using the remaining data as training data. As a result, a new model approximated to the actual dispatching rule can be created with less effect of the deviation from the rule in the production records.” (para. 0023).
US 20210224286 A1 (Wu). “In this way, the original search click data is filtered to remove some data that adversely affects the training of the similarity model, and a relatively reliable initial training sample set may be obtained from a large quantity of original search click records.” (para. 0163).
US 11232369 B1 (Li). “Thus, methods of improving non-spam training data quality and removing spam samples mixed in that non-spam data are required in order to improve the training data and build more reliable classifiers.” (col. 4, lines 58-61).
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MICHAEL C LEE whose telephone number is (571)272-4933. The examiner can normally be reached M-F 12:00 pm - 8:00 pm ET.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Omar Fernandez Rivas can be reached at 571-272-2589. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/MICHAEL C. LEE/Examiner, Art Unit 2128
Prosecution Insights