DETAILED ACTION
The amendments filed 4/13/2026 have been entered. Claims 4, 7, 17, and 19 have been amended. Claims 1-20 remain pending in the application and are discussed on the merits below.
Notice of Pre-AIA or 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 Arguments
Applicant's arguments filed 4/13/2026 have been fully considered but they are not persuasive.
Applicant asserts “the cited portions of the proposed Saito-Liu-Cheung combination do not disclose, teach, or suggest” the features of claim 1 in pages 10-16 of Applicant’s Remarks. Applicant further asserts “No reference, alone or in combination, teaches generating predicted load information in frequency domain from operating condition parameters and converting to time domain for fatigue life determination as claimed” in pages 11-12 of Applicant’s Remarks. However, Examiner respectfully disagrees. Saito teaches the generating predicted load information, Liu teaches using artificial neural networks (ANN) to determine loads, and Cheung teaches an ANN that outputs in a frequency domain and converting the results from the frequency domain to the time domain. The combination of Saito, Liu, and Cheung teaches the limitations of claim 1.
Applicant further asserts “the proposed Saito-Liu-Cheung combination is improper” in pages 12-15 of Applicant’s Remarks. However, Examiner respectfully disagrees. Liu is being relied upon to teach that ANNs can be used to determine loads from flight state and control system parameters and Cheung is brought in to teach that ANNs estimating loads of helicopters can include outputs in a frequency domain and a conversion from the frequency domain into a time domain. Liu further teaches that there are works that analyze signals in the frequency domain. As cited in the Office Action dated 1/12/2026, it would have been obvious to One having ordinary skill in the art to modify Saito’s fatigue estimation with Liu’s neural network to determine loads from parameters in order to use artificial neural networks “to reliably estimate airframe loads on rotary-wing aircraft” (see Liu page 13, paragraph 1), and add the use of an ANN that outputs in a frequency domain and converts to a time domain in order “to verify the outputs from the network” (see page 10, paragraph 2).
Applicant further asserts the independent claim 7 is allowable because of the amended limitation and that “nowhere do the cited portions of Saito disclose, teach or suggest” the new amended limitation. However, Examiner respectfully disagrees. Saito discloses a graph with a fatigue life based on stress and number of cycles to failure (Saito Figure 3). Based on the stress and number of cycles, the Sn curve expresses the remaining fatigue life of the predetermined area in consideration of the damage that the area has received thus far (Saito [0047]-[0049]). Therefore, Saito discloses Applicant’s newly amended limitations in claim 7.
Upon further consideration, Examiner finds Applicant’s arguments regarding claim 17 on pages 17-19 persuasive and claim 17 has been indicated allowable below.
Response to Amendment
Regarding the rejections under 35 USC §103, amendments made to the claims fail to overcome the rejections. The rejections under 35 USC §103 are maintained as outlined below.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
Claims 1, 3, and 6 are rejected under 35 U.S.C. 103 as being unpatentable over Saito et al. (U.S. Patent Application Publication No. 2021/0130012 A1; hereinafter Saito) in view of Liu et al. (“Use of Artificial Neural Networks for Helicopter Load Monitoring”; see reference U on PTO-892; hereinafter Liu) and further in view of Cheung et al. (“Helicopter Loads Synthesis using a Neural Network”; see reference V on PTO-892; hereinafter Cheung).
Regarding claim 1, Saito discloses:
a rotorcraft component (fuselage of an aircraft, see at least [0010]);
a plurality of sensors (acceleration sensor, strain gauge, and the like are installed in aircraft 1, see at least [0035]);
one or more non-transitory computer storage media storing programming for execution by one or more processors (read only memory for storing programs executed by CPU, see at least [0040]), the programming comprising instructions to:
access actual operating condition information for operating condition parameters of the rotorcraft, the actual operating condition information corresponding to sensor measurements from the plurality of sensors and associated with the rotorcraft component (operation data is information recorded in flight recorder and includes history information such as altitude, temperature, and wind speed, see at least [0032]; structural health monitoring (SMH) obtains structural condition data such as damage occurrence, strain condition, and degree of corrosion, see at least [0035]);
analyze, the actual operating condition information to generate initial predicted load information for the rotorcraft component (operation data and monitoring data such as strain is acquired to estimate load, see at least [0065] and Fig. 4)
determine, according to the predicted load information, an estimated fatigue life for the rotorcraft component that is individualized for the rotorcraft component (fatigue life is predicted based on the damage rule, estimated stress, and estimated load, see at least [0066] and Fig. 4).
Saito does not explicitly disclose:
A rotorcraft, comprising:
using an artificial intelligence model
the initial predicted load information comprising a first output data signal in a frequency domain
convert the initial predicted load information to a time domain to generate predicted load information for the rotorcraft component by applying a transform operation to the first output data signal to generate a second output data signal in a time domain
However, Liu teaches:
A rotorcraft (helicopter, see title)
analyze, using an artificial intelligence model, the actual operating condition information to generate initial predicted load information for the rotorcraft component (utilizing multi-layer artificial neural networks (ANNs) to determine airframe loads at fixed locations from flight state and control system parameters obtained during a Black Hawk flight load survey, see at least abstract; ANNs trained using data from steady state flight condition, see at least page 7, paragraph 3)
initial predicted load information to a time domain to generate predicted load information for the rotorcraft (control system parameters analyzed in time domain, see at least page 3, paragraph 4)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito by adding the neural networks taught by Liu with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification for “the strong potential for artificial neural networks to reliably estimate airframe loads on rotary-wing aircraft” (see page 13, paragraph 1).
Additionally, Cheung teaches:
using an artificial intelligence model (ANN output see at least page 11),
the initial predicted load information comprising a first output data signal in a frequency domain (training result showing ANN output in frequency domain, see at least Fig. 9)
convert the initial predicted load information to a time domain to generate predicted load information for the rotorcraft component by applying a transform operation to the first output data signal to generate a second output data signal in a time domain (ANN output transformed back to time domain, see at least Fig. 10)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito and the neural networks and analysis in time domain taught by Liu by adding the domain conversions taught by Cheung with a reasonable expectation of success. Although the ANN outputs are in the frequency domain in Cheung, the analysis of Liu is done in the time domain. The transformation of data from a frequency domain to a time domain (and vice versa) using a Fast Fourier Transform (FFT) function can be done as a design choice or for specific purposes of the data being observed. One of ordinary skill in the art would have been motivated to make this modification in order “to verify the outputs from the network” (see page 10, paragraph 2).
Regarding claim 3, the combination of Saito, Liu, and Cheung teaches the elements above and Saito further discloses:
the rotorcraft component comprises: a pitch link of the rotorcraft; a fuselage of the rotorcraft (fuselage of an aircraft, see at least [0010]); or one or more rotor blades of the rotorcraft.
Regarding claim 6, the combination of Saito, Liu, and Cheung teaches the elements above and Saito further discloses:
the programming further comprises instructions to generate, according to the estimated fatigue life for the rotorcraft component, a maintenance plan that comprises an individualized maintenance recommendation for the rotorcraft component (repair method corresponding to the current degree of damage is presented, see at least [0068]).
Claims 7, 9, and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Saito in view of Liu.
Regarding claim 7, Saito discloses:
A method, comprising: accessing, by a processing device (programs executed by CPU, see at least [0040]), actual operating condition information for an operating condition parameter associated with actual operation of a vehicle, the actual operating condition information corresponding to sensor measurements associated with a vehicle component of the vehicle (operation data is information recorded in flight recorder and includes history information such as altitude, temperature, and wind speed, see at least [0032]; structural health monitoring (SMH) obtains structural condition data such as damage occurrence, strain condition, and degree of corrosion, see at least [0035]);
analyzing, by the processing device, the actual operating condition information to generate predicted load information for the vehicle component (operation data and monitoring data such as strain is acquired to estimate load, see at least [0065] and Fig. 4); and
determining, according to the predicted load information, an estimated fatigue life for the vehicle component that is individualized for the vehicle component (fatigue life is predicted based on the damage rule, estimated stress, and estimated load, see at least [0066] and Fig. 4)
wherein determining, according to the predicted load information, the estimated fatigue life for the vehicle component comprises determining the estimated fatigue life for the vehicle component from estimated fatigue life mappings according to the predicted load information, the estimated fatigue life mappings mapping particular values for predicted load information to particular fatigue life estimations (see Fig. 3; fatigue life can be determined based on the stress, see at least [0047] and [0049])
Saito does not explicitly disclose:
using an artificial intelligence model
However, Liu teaches:
analyze, using an artificial intelligence model, the actual operating condition information to generate initial predicted load information for the rotorcraft component (utilizing multi-layer artificial neural networks (ANNs) to determine airframe loads at fixed locations from flight state and control system parameters obtained during a Black Hawk flight load survey, see at least abstract; ANNs trained using data from steady state flight condition, see at least page 7, paragraph 3)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito by adding the neural networks taught by Liu with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification for “the strong potential for artificial neural networks to reliably estimate airframe loads on rotary-wing aircraft” (see page 13, paragraph 1).
Regarding claim 9, the combination of Saito and Dempsey teaches the elements above and Saito further discloses:
the vehicle is an aircraft (aircraft, see at least [0010]);
the operating condition parameter comprises a flight parameter associated with an actual flight of the aircraft (operation data is information recorded in flight recorder and includes history information such as altitude, temperature, and wind speed, see at least [0032]); and
the actual operating condition information associated with the operating condition parameter comprises a data signal for the flight parameter (structural health monitoring (SMH) obtains structural condition data such as damage occurrence, strain condition, and degree of corrosion, see at least [0035]).
Regarding claim 16, the combination of Saito and Dempsey teaches the elements above and Saito further discloses:
generating, according to the estimated fatigue life for the vehicle component, a maintenance plan that comprises an individualized maintenance recommendation for the vehicle component (repair method corresponding to the current degree of damage is presented, see at least [0068])
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Saito in view of Liu as applied to claim 7 above and further in view of Cheung.
Regarding claim 8, the combination of Saito and Liu teaches the elements above and Saito further discloses:
the sensor measurements are part of a data signal received from a sensor associated with the vehicle component (acceleration sensor, strain gauge, and the like are installed in aircraft 1, see at least [0035]) (acceleration sensor, strain gauge, and the like are installed in aircraft 1, see at least [0035])
Saito does not explicitly disclose:
the method further comprises performing pre-processing on the data signal to generate the actual operating condition information
However, Dempsey teaches:
the method further comprises performing pre-processing on the data signal to generate the actual operating condition information (parameter signal were transformed from time domain to frequency domain, see at least page 5, paragraph 2)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito by adding the domain conversions taught by Cheung with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification “to perform the analyses in the frequency domain” because “the parameter signals are periodic” (see page 5, paragraph 2).
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Saito in view of Liu and Cheung as applied to claim 1, and further in view of Huang et al. (“Convolutional Networks with Dense Connectivity”; see reference V in PTO-892; hereinafter Huang).
Regarding claim 2, the combination of Saito, Liu, and Cheung teaches the elements above but does not teach:
the artificial intelligence model comprises a plurality of layers arranged in a processing sequence, the plurality of layers comprising: at least one first dense layer; at least one second dense layer; and
between the at least one first dense layer and the at least one second dense layer: at least one convolutional layer; at least one normalization layer; and at least one pooling layer
However, Huang teaches:
the artificial intelligence model comprises a plurality of layers arranged in a processing sequence, the plurality of layers comprising: at least one first dense layer; at least one second dense layer; and between the at least one first dense layer and the at least one second dense layer: at least one convolutional layer; at least one normalization layer; and at least one pooling layer (deep DenseNet comprises a dense block 1 with convolution and pooling layers in succession, then a dense block 2, see at least Fig. 2)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito, the neural networks taught by Liu, and the domain conversions taught by Cheung by adding the deep DenseNet taught by Huang with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification “To further improve the information flow between layers” (see page 3, col. 2, para. 2) and to “substantially improve parameter efficiency” (see abstract).
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Saito in view of Liu as applied to claim 7 above, and further in view of Huang.
Regarding claim 10, the combination of Saito and Liu teaches the elements above but does not teach:
the artificial intelligence model comprises a plurality of layers arranged in a processing sequence, the plurality of layers comprising: at least one first dense layer; at least one second dense layer; and between the at least one first dense layer and the at least one second dense layer: at least one convolutional layer; at least one normalization layer; and at least one pooling layer.
However, Huang teaches:
the artificial intelligence model comprises a plurality of layers arranged in a processing sequence, the plurality of layers comprising: at least one first dense layer; at least one second dense layer; and between the at least one first dense layer and the at least one second dense layer: at least one convolutional layer; at least one normalization layer; and at least one pooling layer (deep DenseNet comprises a dense block 1 with convolution and pooling layers in succession, then a dense block 2, see at least Fig. 2)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito and the neural networks taught by Liu by adding the deep DenseNet taught by Huang with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification “To further improve the information flow between layers” (see page 3, col. 2, para. 2) and to “substantially improve parameter efficiency” (see abstract).
Claims 4-5 are rejected under 35 U.S.C. 103 as being unpatentable over Saito in view of Liu and Cheung as applied to claim 1 above, and further in view of Polanco (“Estimation of Structural Component Loads in Helicopters: A Review of Current Methodologies”; already made of record on IDS dated 10/26/2023).
Regarding claim 4, the combination of Saito, Liu, and Cheung teaches the elements above but does not teach:
the artificial intelligence model has been trained according to a training phase that comprises: accessing training data for a time window, the training data comprising: test operating condition information for a test operating condition of a test rotorcraft, the test operating condition information corresponding to sensor measurements associated with a rotorcraft component of the rotorcraft, the test operating condition corresponding to the operating condition parameters of the rotorcraft; and test load information for a training operating condition of one or more test rotorcraft; and training the artificial intelligence model using the training data.
However, Liu teaches:
the artificial intelligence model has been trained according to a training phase (training stage in ANN simulation, see at least page 4, paragraph 3) that comprises:
test operating condition information for a test operating condition of a test vehicle (trained using data from steady state flight condition for Black Hawk, see at least page 7, Fig. 3 and paragraph 3), the test operating condition information corresponding to sensor measurements associated with a vehicle component of the vehicle (strain data gathered from strain gauges, and other sensors captured flight state and control system parameters for test data, see at least page 3, paragraphs 3-6), the test operating condition corresponding to the operating condition parameters of the vehicle (flight state and control system parameters, see at least page 3, paragraph 3); and test load information for a training operating condition of one or more test vehicle (data obtained from flight loads survey for Black Hawk, see at least page 3, paragraph 5); and
training the artificial intelligence model using the training data (training the ANNs was carried out until validation error reached a minimum, see at least page 4, paragraph 4 and page 6, paragraph 3).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito by adding the neural networks taught by Liu with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification for “the strong potential for artificial neural networks to reliably estimate airframe loads on rotary-wing aircraft” (see page 13, paragraph 1).
Additionally, Polanco teaches:
the artificial intelligence model has been trained according to a training phase that comprises: accessing training data for a time window (database generated from helicopter consisting of time histories of 24 flight parameters and 17 loads, see at least page 19, para. 2), the training data comprising:
test operating condition information for a test operating condition of a test rotorcraft, the test operating condition information corresponding to sensor measurements associated with a rotorcraft component of the rotorcraft, the test operating condition corresponding to the operating condition parameters of the rotorcraft (chosen parameters were pitch, roll, and yaw rates, vertical, lateral, and longitudinal accelerations, and longitudinal and lateral control positions, see at least page 19, para. 2, estimate time-varying oscillation of tailboom bending load and pitch link load, see at least page 19, para. 4); and
test load information for a training operating condition of one or more test rotorcraft (testing was conducted on data from seventh maneuver to test estimation accuracy, see at least page 19, para. 4); and
training the artificial intelligence model using the training data (load and flight parameters used to determining weighting coefficients of the neural network, see at least page 18, para. 3; network was trained, see at least page 18, para. 4).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito, the neural networks taught by Liu, and the domain conversions taught by Cheung by adding the neural network trained using a database taught by Polanco with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification for “accuracy in the neural network identification of mean and oscillatory components” (see page 18, para. 3).
Regarding claim 5, the combination of Saito, Liu, Cheung, and Polanco teaches the elements above but Saito does not disclose:
the training data for the time window is in a time domain; and the training phase further comprises converting, prior to training the artificial intelligence model using the training data, at least a portion of the training data to a frequency domain, such that the artificial intelligence model is trained in the frequency domain.
However, Cheung teaches:
the training data for the time window is in a time domain (parameter signals transformed from time domain, see at least page 5, paragraph 2); and
the training phase further comprises converting, prior to training the artificial intelligence model using the training data, at least a portion of the training data to a frequency domain, such that the artificial intelligence model is trained in the frequency domain (parameter signals transformed from time domain to frequency domain using FFT function, see at least page 5, paragraph 2)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito and the neural networks taught by Liu by adding the domain conversions taught by Cheung with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification “since the parameter signals are periodic… to perform the analyses in the frequency domain” (see page 2, paragraph 2).
Claims 11, 13, and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Saito in view of Liu as applied to claims 7 and 17 above, and further in view of Polanco (“Estimation of Structural Component Loads in Helicopters: A Review of Current Methodologies”; already made of record on IDS dated 10/26/2023)
Regarding claim 11, the combination of Saito and Liu teaches the elements above but Saito does not disclose:
the artificial intelligence model has been trained according to training data that comprises: test operating condition information associated with a test operating condition of a test vehicle, the test operating condition information corresponding to sensor measurements associated with a vehicle component of the test vehicle under the test operating condition; and one or more test load signals
However, Liu teaches:
the artificial intelligence model has been trained according to training data (training stage in ANN simulation, see at least page 4, paragraph 3) that comprises:
test operating condition information associated with a test operating condition of a test vehicle (trained using data from steady state flight condition for Black Hawk, see at least page 7, Fig. 3 and paragraph 3), the test operating condition information corresponding to sensor measurements associated with a vehicle component of the test vehicle under the test operating condition (strain data gathered from strain gauges, and other sensors captured flight state and control system parameters for test data, see at least page 3, paragraphs 3-6); and one or more test load signals (data obtained from flight loads survey for Black Hawk, see at least page 3, paragraph 5); and
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito by adding the neural networks taught by Liu with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification for “the strong potential for artificial neural networks to reliably estimate airframe loads on rotary-wing aircraft” (see page 13, paragraph 1).
Additionally, Polanco teaches:
the artificial intelligence model has been trained according to training data (load and flight parameters used to determining weighting coefficients of the neural network, see at least page 18, para. 3; network was trained, see at least page 18, para. 4) that comprises:
test operating condition information associated with a test operating condition of a test vehicle, the test operating condition information corresponding to sensor measurements associated with a vehicle component of the test vehicle under the test operating condition; and one or more test load signals (chosen parameters were pitch, roll, and yaw rates, vertical, lateral, and longitudinal accelerations, and longitudinal and lateral control positions, see at least page 19, para. 2, estimate time-varying oscillation of tailboom bending load and pitch link load, see at least page 19, para. 4; testing was conducted on data from seventh maneuver to test estimation accuracy, see at least page 19, para. 4)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito and the neural networks taught by Liu by adding the neural network trained using a database taught by Polanco with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification for “accuracy in the neural network identification of mean and oscillatory components” (see page 18, para. 3).
Regarding claim 13, the combination of Saito and Liu teaches the elements above but Saito does not disclose:
executing a training phase for training the artificial intelligence model, the training phase comprising: accessing training data for a time window, the training data comprising: test operating condition information for a test operating condition of a test vehicle, the test operating condition information corresponding to sensor measurements associated with a vehicle component of the vehicle, the test operating condition corresponding to the actual operating condition of the vehicle; and test load information for the time window; and training the artificial intelligence model using the training data.
However, Liu teaches:
executing a training phase for training the artificial intelligence model, the training phase comprising (training stage in ANN simulation, see at least page 4, paragraph 3):
test operating condition information for a test operating condition of a test vehicle (trained using data from steady state flight condition for Black Hawk, see at least page 7, Fig. 3 and paragraph 3), the test operating condition information corresponding to sensor measurements associated with a vehicle component of the vehicle (strain data gathered from strain gauges, and other sensors captured flight state and control system parameters for test data, see at least page 3, paragraphs 3-6), the test operating condition corresponding to the actual operating condition of the vehicle (flight state and control system parameters, see at least page 3, paragraph 3); test load information (data obtained from flight loads survey for Black Hawk, see at least page 3, paragraph 5); and
training the artificial intelligence model using the training data (training the ANNs was carried out until validation error reached a minimum, see at least page 4, paragraph 4 and page 6, paragraph 3).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito by adding the neural networks taught by Liu with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification for “the strong potential for artificial neural networks to reliably estimate airframe loads on rotary-wing aircraft” (see page 13, paragraph 1).
Additionally, Polanco teaches:
executing a training phase for training the artificial intelligence model, the training phase comprising: accessing training data for a time window (database generated from helicopter consisting of time histories of 24 flight parameters and 17 loads, see at least page 19, para. 2), the training data comprising:
test operating condition information for a test operating condition of a test vehicle, the test operating condition information corresponding to sensor measurements associated with a vehicle component of the vehicle, the test operating condition corresponding to the actual operating condition of the vehicle (chosen parameters were pitch, roll, and yaw rates, vertical, lateral, and longitudinal accelerations, and longitudinal and lateral control positions, see at least page 19, para. 2, estimate time-varying oscillation of tailboom bending load and pitch link load, see at least page 19, para. 4); and
test load information for the time window (testing was conducted on data from seventh maneuver to test estimation accuracy, see at least page 19, para. 4); and
training the artificial intelligence model using the training data (load and flight parameters used to determining weighting coefficients of the neural network, see at least page 18, para. 3; network was trained, see at least page 18, para. 4).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito and the neural networks taught by Liu by adding the neural network trained using a database taught by Polanco with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification for “accuracy in the neural network identification of mean and oscillatory components” (see page 18, para. 3).
Regarding claim 15, the combination of Saito, Liu, and Polanco teaches the elements above but Saito does not disclose:
the training data is collected from a plurality of test vehicle operations, the test vehicle operations being actual vehicle operations or simulated vehicle operations.
However, Polanco teaches:
the training data is collected from a plurality of test vehicle operations, the test vehicle operations being actual vehicle operations or simulated vehicle operations (database generated from helicopter consisting of time histories of 24 flight parameters and 17 loads, see at least page 19, para. 2)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito and the neural networks taught by Liu by adding the neural network trained using a database taught by Polanco with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification for “accuracy in the neural network identification of mean and oscillatory components” (see page 18, para. 3).
Regarding claim 19, the combination of Saito and Liu teaches the elements above but Saito does not disclose:
the training phase comprises: accessing training data for a time window, the training data comprising: the test operating condition information, the test operating condition information being for a test operating condition of a test vehicle, the test operating condition information corresponding to sensor measurements associated with a vehicle component of the vehicle, the test operating condition corresponding to the operating condition parameter of the vehicle; and the test load information, the test load information being for a training operating condition of one or more test vehicle; and training the artificial intelligence model using the training data.
However, Liu teaches:
the training phase (training stage in ANN simulation, see at least page 4, paragraph 3) comprises:
the test operating condition information, the test operating condition information being for a test operating condition of a test vehicle (trained using data from steady state flight condition for Black Hawk, see at least page 7, Fig. 3 and paragraph 3), the test operating condition information corresponding to sensor measurements associated with a vehicle component of the vehicle (strain data gathered from strain gauges, and other sensors captured flight state and control system parameters for test data, see at least page 3, paragraphs 3-6), the test operating condition corresponding to the operating condition parameter of the vehicle (flight state and control system parameters, see at least page 3, paragraph 3); and the test load information, the test load information being for a training operating condition of one or more test vehicle (data obtained from flight loads survey for Black Hawk, see at least page 3, paragraph 5); and
training the artificial intelligence model using the training data (training the ANNs was carried out until validation error reached a minimum, see at least page 4, paragraph 4 and page 6, paragraph 3).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito by adding the neural networks taught by Liu with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification for “the strong potential for artificial neural networks to reliably estimate airframe loads on rotary-wing aircraft” (see page 13, paragraph 1).
Additionally, Polanco teaches:
the artificial intelligence model has been trained according to a training phase that comprises: accessing training data for a time window (database generated from helicopter consisting of time histories of 24 flight parameters and 17 loads, see at least page 19, para. 2), the training data comprising:
the test operating condition information, the test operating condition information being for a test operating condition of a test vehicle, the test operating condition information corresponding to sensor measurements associated with a vehicle component of the vehicle, the test operating condition corresponding to the operating condition parameter of the vehicle (chosen parameters were pitch, roll, and yaw rates, vertical, lateral, and longitudinal accelerations, and longitudinal and lateral control positions, see at least page 19, para. 2, estimate time-varying oscillation of tailboom bending load and pitch link load, see at least page 19, para. 4); and
the test load information, the test load information being for a training operating condition of one or more test vehicle (testing was conducted on data from seventh maneuver to test estimation accuracy, see at least page 19, para. 4); and
training the artificial intelligence model using the training data (load and flight parameters used to determining weighting coefficients of the neural network, see at least page 18, para. 3; network was trained, see at least page 18, para. 4).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito and the neural networks taught by Liu by adding the neural network trained using a database taught by Polanco with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification for “accuracy in the neural network identification of mean and oscillatory components” (see page 18, para. 3).
Claim 14 is rejected under 35 U.S.C. 103 as being unpatentable over Saito in view of Liu and Polanco as applied to claim 13 above, and further in view of Cheung.
Regarding claim 14, the combination of Saito, Liu, and Polanco teaches the elements above but Saito does not disclose:
the training data for the time window is in a time domain; and the method further comprises converting, prior to training the artificial intelligence model using the training data, at least a portion of the training data to a frequency domain, such that the artificial intelligence model is trained in the frequency domain.
However, Cheung teaches:
the training data for the time window is in a time domain (parameter signals transformed from time domain, see at least page 5, paragraph 2); and
the training phase further comprises converting, prior to training the artificial intelligence model using the training data, at least a portion of the training data to a frequency domain, such that the artificial intelligence model is trained in the frequency domain (parameter signals transformed from time domain to frequency domain using FFT function, see at least page 5, paragraph 2)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito and the neural networks taught by Liu by adding the domain conversions taught by Cheung with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification “since the parameter signals are periodic… to perform the analyses in the frequency domain” (see page 2, paragraph 2).
Claim 12 is rejected under 35 U.S.C. 103 as being unpatentable over Saito in view of Liu and Polanco as applied to claim 11 above, and further in view of Cheung and Giering et al. (U.S. Patent Application Publication No. 2018/0217585 A1; hereinafter Giering).
Regarding claim 12, the combination of Saito, Liu, and Polanco teaches the elements above and Saito further discloses:
analyzing, the actual operating condition information to generate the predicted load information for the vehicle component comprises generating, by the processing device (operation data and monitoring data such as strain is acquired to estimate load, see at least [0065] and Fig. 4)
Saito does not disclose:
the artificial intelligence model has been trained in a frequency domain;, initial predicted load information in the frequency domain; and the method further comprises converting, by the processing device, the initial predicted load information to a time domain.
However, Cheung teaches:
initial predicted load information in the frequency domain (training result showing ANN output in frequency domain, see at least Fig. 9); and the method further comprises converting, by the processing device, the initial predicted load information to a time domain (ANN output transformed back to time domain, see at least Fig. 10)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito, the neural networks taught by Liu, and the neural network trained using a database taught by Polanco by adding the domain conversions taught by Cheung with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order “to verify the outputs from the network” (see page 10, paragraph 2).
Additionally, Giering teaches:
the artificial intelligence model has been trained in a frequency domain (prognostic and health monitoring (PHM) sensors are converted into frequency domain data and are labeled to use as PHM training indicators, see at least [0003]);
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the fatigue estimation disclosed by Saito, the neural networks taught by Liu, the neural network trained using a database taught by Polanco, and the domain conversions taught by Cheung by adding the training using frequency domain data taught by Giering with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification for “creation and use of monitoring models from multiple sensor inputs for health and prognostic monitoring” (see [0014]).
Allowable Subject Matter
Claims 17-20 are allowed.
The following is an examiner’s statement of reasons for allowance:
Claim 17 is allowable for reciting “analyze, using an artificial intelligence model configured to determine predicted structural loads from operating condition information based on training using training data that includes test operating condition information and corresponding test load information, the actual operating condition information to generate initial predicted load information for the vehicle component in a frequency domain, the artificial intelligence model having been trained according to a training phase that comprises converting the training data from a time domain to the frequency domain and training the artificial intelligence model using the training data in the frequency domain; convert the initial predicted load information from the frequency domain to the time domain to generate predicted load information for the vehicle component; and determine, according to the predicted load information in the time domain, an estimated fatigue life for the vehicle component that is individualized for the vehicle component” in combination with the other elements of claim 17.
Saito discloses structural health monitoring of an aircraft by obtaining structural condition data such as damage occurrence, strain condition, and a degree of corrosion an using operation and monitoring data, such as strain, to estimate load. The fatigue life of the area of the aircraft is based on the damage, estimated stress, and estimated load. However, Saito does not disclose using a trained artificial intelligence model to determined predicted structural loads, a frequency domain, and a time domain.
Liu teaches a multi-layer artificial neural network (ANNs) to determine airframe loads at fixed locations from flight state and control system parameters obtained during a Black Hawk flight load survey. The ANNs having been trained using data from a steady flight condition. Liu does not teach converting data from a time domain to a frequency domain and converting from a frequency domain to a time domain.
Cheung teaches pre-processing data for input to a neural network wherein the data is transformed from a time domain to a frequency domain and training an ANN using the data in the frequency domain and converting the data output from the frequency domain into a time domain. However, Cheung does not teach “analyze, using an artificial intelligence model configured to determine predicted structural loads from operating condition information based on training using training data that includes test operating condition information and corresponding test load information, the actual operating condition information to generate initial predicted load information for the vehicle component in a frequency domain,” “the artificial intelligence model having been trained according to a training phase that comprises converting the training data from a time domain to the frequency domain,” and “determine, according to the predicted load information in the time domain, an estimated fatigue life for the vehicle component that is individualized for the vehicle component.”
Giering teaches prognostic and health monitoring sensors are converted into frequency domain data and labeled to use as training indicators. However, Giering does not teach “analyze, using an artificial intelligence model configured to determine predicted structural loads from operating condition information based on training using training data that includes test operating condition information and corresponding test load information, the actual operating condition information to generate initial predicted load information for the vehicle component in a frequency domain, the artificial intelligence model having been trained according to a training phase that comprises converting the training data from a time domain to the frequency domain and training the artificial intelligence model using the training data in the frequency domain; convert the initial predicted load information from the frequency domain to the time domain to generate predicted load information for the vehicle component; and determine, according to the predicted load information in the time domain, an estimated fatigue life for the vehicle component that is individualized for the vehicle component.”
Polanco teaches a database generated from helicopter time histories of flight parameters and loads and a time-varying oscillation of tail boom bending load and pitch link load being estimated. Testing was conducted on data to test estimation accuracy. However, Polanco does not teach “analyze, using an artificial intelligence model configured to determine predicted structural loads from operating condition information based on training using training data that includes test operating condition information and corresponding test load information, the actual operating condition information to generate initial predicted load information for the vehicle component in a frequency domain, the artificial intelligence model having been trained according to a training phase that comprises converting the training data from a time domain to the frequency domain and training the artificial intelligence model using the training data in the frequency domain; convert the initial predicted load information from the frequency domain to the time domain to generate predicted load information for the vehicle component; and determine, according to the predicted load information in the time domain, an estimated fatigue life for the vehicle component that is individualized for the vehicle component.”
The combination of Saito, Liu, Cheung, Geiring, and Polanco fail to teach “analyze, using an artificial intelligence model configured to determine predicted structural loads from operating condition information based on training using training data that includes test operating condition information and corresponding test load information, the actual operating condition information to generate initial predicted load information for the vehicle component in a frequency domain, the artificial intelligence model having been trained according to a training phase that comprises converting the training data from a time domain to the frequency domain and training the artificial intelligence model using the training data in the frequency domain; convert the initial predicted load information from the frequency domain to the time domain to generate predicted load information for the vehicle component; and determine, according to the predicted load information in the time domain, an estimated fatigue life for the vehicle component that is individualized for the vehicle component” in combination with the other elements of claim 7. Furthermore, it would not have been obvious to one having ordinary skill in the art to combine Saito, Liu, Cheung, Geiring, and Polanco to achieve the newly added limitations (underlined).
Claims 18-20 are allowable because they are dependent on claim 17 and further define and limit the claims.
Any comments considered necessary by applicant must be submitted no later than the payment of the issue fee and, to avoid processing delays, should preferably accompany the issue fee. Such submissions should be clearly labeled “Comments on Statement of Reasons for Allowance.”
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.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HANA LEE whose telephone number is (571)272-5277. The examiner can normally be reached Monday-Friday: 7:30AM-4:30PM EST.
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, Jelani Smith can be reached at (571) 270-3969. 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.
/H.L./Examiner, Art Unit 3662
/DALE W HILGENDORF/Primary Examiner, Art Unit 3662
Prosecution Insights