DETAILED ACTION
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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 11/24/2025 was in compliance with the provisions of 37 CFR 1.97. Accordingly, the IDS is being considered by the examiner.
Response to Amendment
Claims 1, 3, 17-18, 20, and 34 are amended, claims 2 and 19 are cancelled and claims 35-36 are new. Claims 1, 3-18, and 20-36 are pending.
Response to Arguments
Applicant's arguments filed 11/21/2025 have been fully considered.
Regarding the rejections of independent claims 1 and 18 under 101, Applicant contends on page 16 of the response that the amended claims include additional elements, which considered in combination with the recited judicial exception, integrate the judicial exception into a practical application in terms of an improvement to the technical field of mobility systems and automated systems for monitoring the condition of a mobility system. In support, Applicant notes the recited use of sensor data collected during use of the mobility system and processing the data using a trained machine learning system that has been trained during two distinct training stages.
The Examiner submits that the use of sensors to collect vibration data represents high level data collection having no particularized functional relation to the processing steps except as simply providing the input data such that it does not embody an improvement to any particular technical field including mobility systems and/or equipment monitoring. The processing of the sensor data appears to be implemented by standard computer processing functions/components to simply implement the underlying function (predicting wear and characterizing lifecycle stage) such that no improvement particular to a particular technical field is apparent.
Regarding the training being implemented in two distinct stages, the Examiner notes that the elements characterizing the two-stage training are framed to clearly omit the related data collection and training functions from being implemented by the recited system/method. Instead, these features only characterize the origin of the model that is utilized such that the potential effect of these functions in terms of integrating the judicial exception into a practical application is largely moot. (The claimed two-step training process is not recited as being carried out by the system or as part of the method of the claimed invention, but is only descriptive of the trained machine learning system that is used in the system and method.)
On page 17, Applicant contends without support that the limitations of claims 1 and 18 cannot be practically be performed in the human mind and therefore do not recite an abstract idea in the form of a mental process. The Examiner submits, as explained in further detail in the grounds for rejecting claims 1 and 18, that the steps of “monitoring a condition of a mobility system,” “receive the data,” and “predict wear and tear experienced by the mobility system and characterize a lifecycle stage of the mobility system” “based on at least the received data,” may be performed as mental processes in relatively simple situations and therefore would fall within the mental processes exception; in more complex situations, they would be implemented as mathematical calculations which would fall within the mathematical calculations exception.
Applicant further contends on page 17 of the response that claims 1 and 18 employ significantly more than generic computer components and instead requires that the limitations be performed by a specially adapted controller arrangement that is structured and configured to implement a trained machine learning system that is trained in a novel manner using two distinct stages. Applicant also cites Desjardins as supporting Applicant’s contention that the additional elements integrate the judicial exception into a practical application.
The Examiner submits that the “controller” recited in claim 1 is not characterized in any particular manner other than for performing ordinary processing functions such as implementing the machine learning system by receiving the data (input function), predicting wear and characterizing lifecycle stage (processing the input data using the machine learning system), and generating and transmitting an alert corresponding to the processing results. In this manner, the controller as claimed represents nothing further than a data processing mechanism for implementing the judicial exception as well as ordinary data processing functions and therefore does not represent an improvement in a technical field or use of the judicial exception by a particularized machine.
Regarding Applicant’s reference to Desjardins, the Examiner acknowledges that improvements to artificial intelligence related technology may be eligible under 101. However, as noted above any potential effect of the “training” steps incorporated by amendment into claims 1 and 18 in terms of integration into a practical application are moot because the claims system and method do not positively recite the training steps as being performed by the recited system/method.
Regarding USPTO Step 2B, Applicant contends on page 18 of the response that the limitations of claims 1 and 18 are patentably distinct over the prior art and therefore represent an “unconventional” solution that involves more than well-understood, routine, and conventional activities previous known to the industry. Applicant’s arguments relating to inventiveness are set forth on pages 19-20 in relation to the 103 rejection and are directed to the “training” steps. As noted above, neither of claims 1 and 18 positively recites these steps as being performed by the recited system/method. Furthermore, the Examiner submits that as set forth in the current grounds for rejecting claims 1 and 18 under 103, that these steps are unpatentable under 103 over Claessens (US 2021/0335061 A1) in view of Turturiello (US 2004/0262859 A1), and in further view of Mhatre A, Reese N, Pearlman J (2020) Design and evaluation of a laboratory-based wheelchair castor testing protocol using community data. PLoS ONE 15(1): e0226621, (Mhatre) as provided by Applicant.
Regarding the rejections of independent claims 1 and 18 under 103, Applicant contends that Claessens does not teach "wherein the trained machine learning system has been previously trained in two stages to predict wear and tear experienced by the mobility system during use and failure conditions based on the data indicative of use of the mobility system, the two stages including (i) a first stage wherein first training data comprising shock data is collected while a first mobility system is tested on a mechanical test bed until a prominent failure mode occurs and wherein a base machine learning model is trained using the first training data, and (ii) a second stage wherein second training data comprising road shock data and environmental data is collected while a plurality of second mobility systems are tested in actual use conditions until the prominent failure mode occurs in each of the second mobility systems and wherein the trained base machine learning model is trained using the second training data to create the trained machine learning system."
As set forth in further detail in the current grounds of rejection under 103, the combination of Claessens, Turturiello, and Mhatre teaches or otherwise renders the elements added by amendment obvious.
Claim Objections
Claim 36 is objected to because of the following informalities:
Claim 36 appears to have incorrect claim dependency. It appears that claim 36, which recites “the method,” should depend from claim 18.
Appropriate correction is required.
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, 3-18, and 20-36 are rejected under 35 U.S.C. 101 because the claimed invention in each of these claims is directed to the abstract idea judicial exception without significantly more.
Claim 1, substantially representative also of independent claim 18, recites:
“[a] system for monitoring a condition of a mobility system, comprising:
a number of sensors coupled to the mobility system, the number of sensors being structured and configured to generate data indicative of use of the mobility system during use, wherein the data includes shock and/or vibration data indicative of one or more shocks and/or vibrations experienced by the mobility system using use; and
a controller implementing a trained machine learning system, wherein the trained machine learning system has been previously trained in two stages to predict wear and tear experienced by the mobility system during use and failure conditions based on the data indicative of use of the mobility system, the two stages including: (i) a first stage wherein first training data comprising shock data is collected while a first mobility system is tested on a mechanical test bed until a prominent failure mode occurs and wherein a base machine learning model is trained using the training data, and (ii) a second stage wherein second training data comprising road shock data and environmental data is collected while a plurality of second mobility systems are tested in actual use conditions until the prominent failure mode occurs in each of the second mobility systems and wherein the trained based machine learning model is trained using the second training data to create the trained machine learning system, wherein the controller is structured and configured to:
receive the data;
predict wear and tear experienced by the mobility system and characterize a lifecycle stage of the mobility system using the trained machine learning system based on at least the received data; and
generate and transmit an alert for required maintenance for and/or predicted breakdown of the mobility system based on the characterized lifecycle stage.”
The claim limitations considered to fall within in the abstract idea are highlighted in bold font above and the remaining features are “additional elements.”
Step 1 of the subject matter eligibility analysis entails determining whether the claimed subject matter falls within one of the four statutory categories of patentable subject matter identified by 35 U.S.C. 101: process, machine, manufacture, or composition of matter. Claim 1 recites a system and claim 18 recites a method and each therefore falls within a statutory category.
Step 2A, Prong One of the analysis entails determining whether the claim recites a judicial exception such as an abstract idea. Under a broadest reasonable interpretation, the highlighted portions of claim 13 fall within the abstract idea judicial exception. Specifically, under the 2019 Revised Patent Subject Matter Eligibility Guidance, the highlighted subject matter falls within the mental processes category (including an observation, evaluation, judgment, opinion) and the mathematical concepts category (mathematical relationships, mathematical formulas or equations, mathematical calculations). MPEP § 2106.04(a)(2).
The recited functions:
“monitoring a condition of a mobility system” and
“receive the data; and
predict wear and tear experienced by the mobility system and characterize a lifecycle stage of the mobility system” “based on at least the received data,” may be performed as mental processes.
Monitoring a condition of a mobility system may be performed via mental processes (e.g., evaluation). Receiving data indicative of mobility system use using the data for predicting wear and tear experienced by the mobility system and characterizing a lifecycle stage of the mobility system based on the received data may also be performed via mental processes (e.g., observation of the mobility system use data such as may be provided on a computer display and evaluation and judgement for predicting wear and tear and characterizing a lifecycle stage).
The recited function “predict wear and tear experienced by the mobility system and characterize a lifecycle stage of the mobility system using the trained machine learning system based on at least the received data” is determined by the Examiner as falling within the mathematical relationships sub-category of mathematical concepts (MPEP 2106.04(a)(2)) because as disclosed in Applicant’s specification in paragraph [0030] the machine learning system is implemented as a time-series regression model, which is widely known as being fundamentally characterized by mathematical relations and calculations and therefore constitutes mathematical relationships.
Step 2A, Prong Two of the analysis entails determining whether the claim includes additional elements that integrate the recited judicial exception into a practical application. “A claim that integrates a judicial exception into a practical application will apply, rely on, or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception, such that the claim is more than a drafting effort designed to monopolize the judicial exception” (MPEP § 2106.04(d)).
MPEP § 2106.04(d) sets forth considerations to be applied in Step 2A, Prong Two for determining whether or not a claim integrates a judicial exception into a practical application. Based on the individual and collective limitations of claim 1 and applying a broadest reasonable interpretation, the most applicable of such considerations appear to include: improvements to the functioning of a computer, or to any other technology or technical field (MPEP 2106.05(a)); applying the judicial exception with, or by use of, a particular machine (MPEP 2106.05(b)); and effecting a transformation or reduction of a particular article to a different state or thing (MPEP 2106.05(c)).
Regarding improvements to the functioning of a computer or other technology, none of the “additional elements” including “a number of sensors coupled to the mobility system, the number of sensors being structured and configured to generate data indicative of use of the mobility system during use,” “a controller implementing a trained machine learning system, wherein the trained machine learning system has been previously trained in two stages to predict wear and tear experienced by the mobility system during use and failure conditions based on the data indicative of use of the mobility system, the two stages including: (i) a first stage wherein first training data comprising shock data is collected while a first mobility system is tested on a mechanical test bed until a prominent failure mode occurs and wherein a base machine learning model is trained using the training data, and (ii) a second stage wherein second training data comprising road shock data and environmental data is collected while a plurality of second mobility systems are tested in actual use conditions until the prominent failure mode occurs in each of the second mobility systems and wherein the trained based machine learning model is trained using the second training data to create the trained machine learning system, ” and “generate and transmit an alert for required maintenance for and/or predicted breakdown of the mobility system based on the characterized lifecycle stage,” in any combination appear to integrate the abstract idea in a manner that technologically improves any aspect of a device or system that may be used to implement the highlighted step or a device for implementing the highlighted step such as a signal processing device or a generic computer. Using sensors coupled to a mobility system (a unit for ambulation (e.g., an automobile or wheelchair)) for generating data indicative use of the mobility system represents high-level data gathering that is substantially unrelated to the subsequent data processing and therefore constitutes extra solution activity. A controller implementing a trained machine learning system represents data processing for preparing instructions for implementing the elements entailed in the judicial exception using routine, conventional data processing structure and therefore constitutes extra solution activity. The training process including two training phases in which test bed data is collected in one phase and actual use by multiple mobility systems is collected in the other represents high level data collection. Furthermore, the Examiner notes that these training steps are not positively recited as being functions actually implemented by the recited system (“machine learning system has been previously trained…”). Instead, these steps merely characterize the machine learning system in terms of a process that is performed outside the recited system such that they do not have a significant role in terms of potential integration of the judicial exception into a practical application. Generating and transmitting an alert for required maintenance for and/or predicted breakdown of the mobility system based on the characterized lifecycle stage represents routine, conventional data processing in terms of outputting results that is substantially unrelated to the preceding data processing and therefore constitutes extra solution activity.
Regarding application of the judicial exception with, or by use of, a particular machine, the additional element “a number of sensors coupled to the mobility system, the number of sensors being structured and configured to generate data indicative of use of the mobility system during use,” is broadly recited as configured and implemented as mere data-gathering rather than being a particularized manner of implementing condition monitoring of a mobility system.
Regarding a transformation or reduction of a particular article to a different state or thing, claim 1 does not include any such transformation or reduction. Instead, claim 1 as a whole entails collecting input information (sensors coupled to mobility system and configured to generate usage data), applying standard processing techniques (conventional data processing including application of machine learning) to the information to obtain and deduce lifecycle information of the mobility system with the additional element “generate an alert for required maintenance for and/or predicted breakdown of the mobility system based on the characterized lifecycle stage” failing to provide a meaningful integration of the abstract idea (characterizing a lifecycle stage of the mobility system based on the received data) in an application that transforms an article to a different state. The Examiner notes that even if “using the trained machine learning system” is considered to fall outside the abstract idea exception (i.e., falls outside the mathematical relations exception) and therefore be considered an additional element, this feature represents conventional data processing in terms of executing processing instructions to implement the limitation falling within the abstract idea exception (i.e., characterizing a lifecycle stage of the mobility system), and therefore constitutes extra solution activity.
The additional elements therefore represent extra-solution activity that does not integrate the judicial exception into a practical application. In view of the various considerations encompassed by the Step 2A, Prong Two analysis, claim 1 does not include additional elements that integrate the recited abstract idea into a practical application.
Therefore, claim 1 is directed to a judicial exception and requires further analysis under Step 2B.
Regarding Step 2B, and as explained in the Step 2A Prong Two analysis, the additional elements in claim 1 constitute extra solution activity and therefore do not result in the claim as a whole amounting to significantly more than the judicial exception per Step 2B. Furthermore, the additional elements appear to be generic and well understood as evidenced by the disclosures of Claessens (US 2021/0335061 A1) and Chen (US 2017//0314961 A1), each of which teach substantially similar data processing configuration for collecting and processing sensor data for a mobility system. As explained in the grounds for rejecting claim 1 under 103, Claessens teaches “a number of sensors coupled to the mobility system, the number of sensors being structured and configured to generate data indicative of use of the mobility system during use,” “a controller implementing a trained machine learning system” in which the training includes two phases including a phase in which test bed data is collected and a phase in which actual use data for multiple vehicles is collected, and “generate an alert for required maintenance for and/or predicted breakdown of the mobility system based on the characterized lifecycle stage.” Chen also teaches deploying sensor on mobility systems ([0006], [0083], [0091] sensors deployed in physical systems that may be vehicles/automobiles) and using a controller implemented by conventional data processing means/functions for processing received sensor data to determine lifecycle related information (FIG. 1 Modeling and Analysis System 102 including modeling module 120, analytic engine 122, and anomaly detection module 126 executed by processor 104 to processing input time series data 114 (from sensors) FIG. 2 depicting the processing sequence in which the input time series data is processed and an output alert is generated in block 218), and providing output 128 to implement a control function.
Therefore, the additional elements are insufficient to amount to significantly more than the judicial exception.
Claim 1 is therefore not patent eligible.
Independent claim 18 includes substantially the same elements as claim 1 falling within the abstract idea judicial exception and does not include further additional elements that either integrate the judicial exception into a practical application or result in the claim as a whole amounting to significantly more than the judicial exception.
Claim 18 is therefore also not patent eligible.
Claims 3-17 and 35 depending from claim 1, and claims 20-34 and 36 depending from claim 18 provide additional features/steps which are part of an expanded algorithm that includes the abstract idea of claim 1 (Step 2A, Prong One). None of dependent claims 3-17 and 20-36 recite additional elements that integrate the abstract idea into practical application (Step 2A, Prong Two), and all fail the “significantly more” test under the step 2B for similar reasons as discussed with regards to claim 1.
Claim 3, substantially representative also of claim 20, is an extension of the judicial exception represented by the characterizing a lifecycle state limitation of claim 1 because claim 3 characterizes the nature of the lifecycle stage determination, which may be performed via mental processes.
Claims 4-8, substantially representative also of claims 21-25, include additional elements characterizing the type of sensors used with claims 8 and 25 representing the most limiting combination of sensor types. As recited in claims 4-8 and 21-25, an accelerometer, magnetometer, and gyroscope, even in combination, represent high-level data collection bearing no particularized relation to a particular manner of the subsequent data processing steps and therefore constitute extra solution activity that neither integrates the judicial exception into a practical application nor results in the claim as a whole amounting to significantly more than the judicial exception. Furthermore, using combinations of such sensors is conventional and well known in the art as evidenced by the disclosure of Claessens (see grounds for rejecting claims 4-8 and 21-25 under 103), Hamilton (US 20220142834 A1) ([0024]-[0025] and [0124] disclosing accelerometer, gyroscope, and magnetometer sensors mounted to track and processing data relating to wheelchair kinematics), and Robinson (US 2023/0105957 A1) (FIG. 3 depicting use of a combination of kinematics sensors including an accelerometer and magnetometer combined with a temperature and humidity sensor) such that the additional elements do not result in the claim as a whole amounting to significantly more than the judicial exception.
Claim 9, substantially representative also of claim 26, recites the additional element “a number of environmental conditions sensors coupled to the mobility system, the number of environmental conditions sensors being structured and configured to generate environmental condition data indicative of one or more environmental conditions around the mobility system during use of the mobility system,” which represents high-level data collection using known means (environmental sensors) and therefore constitutes extra solution activity that neither integrates the judicial exception into a practical application nor results in the claim as a whole amounting to significantly more than the judicial exception. Furthermore, including environmental sensor with kinematic sensors (e.g., accelerometers, magnetometers, and/or gyroscopes) for monitoring vehicles is conventional and well known in the art as disclosed by Claessens (see rejections of claims 9 and 26 under 103), Robinson (FIG. 3), and Hamilton ([0124]).
Claims 9 (and similarly claim 26) further recites “wherein the trained machine learning system is structured and configured to receive the environmental condition data and characterize the degree of wear based on the received shock and/or vibration data and the environmental condition data,” which is an extension of the machine learning limitation in claim 1 (falls within the mathematical relationships sub-category of mathematical concepts (MPEP 2106.04(a)(2)) because as disclosed in Applicant’s specification in paragraph [0030] the machine learning system is implemented as a time-series regression model, which is widely known as being fundamentally characterized by mathematical relations and calculations and therefore constitutes mathematical relationships).
Furthermore, even if “wherein the trained machine learning system is structured and configured to receive the environmental condition data and characterize the degree of wear based on the received shock and/or vibration data and the environmental condition data” is considered to fall outside the abstract idea exception (i.e., falls outside the mathematical relations exception), and therefore be considered an additional element, this feature represents conventional data processing in terms of executing processing instructions to implement the limitation falling within the abstract idea exception (i.e., characterizing a degree of wear), and therefore constitutes extra solution activity. More specifically, this limitation provides no particular detail in terms of the function of the machine learning except that it processes the input data presented (combination of environmental and kinematic data) to implement “characterize the degree of wear” which falls within the abstract idea exception because it may be performed via mental processes, such that overall it also constitutes extra solution activity.
Claims 10-11, substantially representative also of claims 27-28, include additional elements characterizing the type of sensors used with claims 11 and 28 representing the most limiting combination of sensor types. As recited in claims 10-11 and 27-28, a temperature sensor and a humidity sensor, even in combination, represent high-level data collection bearing no particularized relation to a particular manner of the subsequent data processing steps and therefore constitute extra solution activity that neither integrates the judicial exception into a practical application nor results in the claim as a whole amounting to significantly more than the judicial exception. Furthermore, using combinations of such sensors and in further combination with kinematics sensors (e.g., accelerometers, magnetometers, and/or gyroscopes) is conventional and well known in the art as evidenced by the disclosure of Claessens (see grounds for rejecting claims 10-11 and 27-28 under 103), Hamilton ([0124], and Robinson (FIG. 3) such that the additional elements do not result in the claim as a whole amounting to significantly more than the judicial exception.
Claim 12, substantially representative also of claim 29, further recites “wherein the trained machine learning system is structured and configured to characterize the degree of wear by determining a usage index based on at least the received shock and/or vibration data,” which is an extension of the machine learning limitation in claim 1 (falls within the mathematical relationships sub-category of mathematical concepts (MPEP 2106.04(a)(2)) because as disclosed in Applicant’s specification in paragraph [0030] the machine learning system is implemented as a time-series regression model, which is widely known as being fundamentally characterized by mathematical relations and calculations and therefore constitutes mathematical relationships).
Furthermore, even if “wherein the trained machine learning system is structured and configured to characterize the degree of wear by determining a usage index based on at least the received shock and/or vibration data” is considered to fall outside the abstract idea exception (i.e., falls outside the mathematical relations exception), and therefore be considered an additional element, this feature represents conventional data processing in terms of executing processing instructions to implement a limitation falling within the abstract idea exception (i.e., determining a usage index based on the received shock and/or vibration data), and therefore constitutes extra solution activity. More specifically, this limitation provides no particular detail in terms of the function of the machine learning except that it processes the input data to implement “determining a usage index” which falls within the abstract idea exception because it may be performed via mental processes, such that overall it also constitutes extra solution activity.
Claims 12 (and similarly claim 29) further recites “wherein the controller is structured and configured to generate the alert when the determined usage index exceeds a certain predetermined threshold value,” which represents routine, convention data output processing (generate an alert) in response to action that may be performed via mental processes (determining that a usage index exceeds a threshold) and therefore constitutes extra solution activity that neither integrates the judicial exception into a practical application nor results in the claim as a whole amounting to significantly more than the judicial exception.
Claim 13, substantially representative also of claim 30, falls within the mathematical relations subcategory of the mathematical concepts exception because Miner’s rule (also known as Palmgren-Miner rule) is fundamentally characterized by mathematical relations and calculations and therefore constitutes mathematical relationships.
Claims 14 and 15, substantially representative of claims 31 and 32, respectively, further recite additional elements that constitute extra solution activity that neither integrates the judicial exception into a practical application nor results in the claim as a whole amounting to significantly more than the judicial exception because these elements characterize relatively arbitrary and otherwise conventional structures for implementing mobility system monitoring that have no significant functional relations to the elements constituting a judicial exception.
Furthermore, the elements in claims 14 and 15 represent conventional and well-understood data processing configuration for monitoring devices such as mobility systems as disclosed by Claessens (see grounds for rejecting claims 14-15 and 31-32 under 103) and Robinson (FIG. 1 multi-sensor node device 140 as part of control function in remote communication with computing system 160 and user device 170 that per [0040] and [0047] also perform control functions; [0007] multi-sensor node device coupled locally (within vehicle that includes component being monitored); FIG. 2 multi-sensor node device including processing/control function).
Claims 16-17, substantially representative also of claims 33-34, fall within the mathematical relations sub-category of the mathematical concepts exception because regression-type machine learning models, including time series regression models, are fundamentally characterized by mathematical relations and calculations and therefore constitutes mathematical relationships.
Dependent claims 3-17 and 20-36 therefore also constitute ineligible subject matter.
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.
Claims 1, 3-16 and 18, 20-33, and 35-36 are rejected under 35 U.S.C. 103 as being unpatentable over Claessens (US 2021/0335061 A1) in view of Turturiello (US 2004/0262859 A1), and in further view of Mhatre A, Reese N, Pearlman J (2020) Design and evaluation of a laboratory-based wheelchair castor testing protocol using community data. PLoS ONE 15(1): e0226621, (Mhatre) as provided by Applicant.
As to claim 1, Claessens teaches “[a] system (FIG. 4 system 400) for monitoring a condition of a” [vehicle] (FIG. 4 system 400 including sensors 406 and monitoring component 430 configured to monitor vehicle 402; [0011] systems for monitoring vehicle health), comprising:
a number of sensors coupled to the” [vehicle] (FIG. 1 sensors 102 coupled to vehicle 100; FIG. 2 sensors 406 integrated with vehicle 402), the number of sensors being structured and configured to generate data indicative of use of the” [vehicle] “during use (sensors 102 in FIG. 1 and sensors 406 incorporated with a vehicle and therefore structured and configured to monitor conditions related to use of the vehicle; [0035]-[0036] sensors including IMU, accelerometers, GPS, electrical parameter sensors configured to generate data indicative of operations/use of the vehicle; [0038], [0041], and [0046] sensor generated data may include dynamic kinematic data such as vibration), wherein the data includes shock and/or vibration data ([0012] sensor data may include acoustic and/or inertial measurements; [0038] sensor data may include vibration) indicative of one or more shocks and/or vibrations experienced by the" [vehicle] “during use ([0038] and [0041] sensed vibration the result of vibration experienced by vehicle (detection of kinematic metric such as vibration would occur during using vehicle internal or external motion)); and
a controller (FIG. 4 vehicle computing device 404 individually and in combination with computing device 440) implementing a trained machine learning system (FIG. 4 vehicle computing device 404 implements monitoring component 430 that per [0063] may be implemented as a machine learning model that has been trained; computing device 440 implements machine learning component 450 and training component 446; [0077]), wherein the trained machine learning system has been previously trained ([0063] machine learning model has been trained)” in two stages ([0106] model training may be continued (model retrained) using additional sensor data) “to predict wear and tear experienced by the mobility system during use and failure conditions based on the data indicative of use of the mobility system (Abstract and [0015] disclosing that the model is trained to identify operating status based on sensor input data; [0011] “operating status” may be component wear and/or likelihood of failure). As interpreted in view of Applicant’s specification, the model “identifying” is entailed within a model “prediction” in terms of identifying what the wear/failure condition is expected to be based on the modeling; FIG. 7 step 708 and 710, [0104]), the two stages including: (i) a first stage wherein first training data comprising” [sensor] “data is collected while a first mobility system is tested on a mechanical test bed ([0099] sensor data may be test data for the vehicle be generated on a test bench; [0101] sensor data may be used as training data) until a prominent failure mode occurs ([0025] training data may be labeled to indicate ground truth operating status such as indication of wear, indication of anomaly (the labeling of ground truth indicates noted occurrence of the label condition such as wear and/or anomaly)) and wherein a base machine learning model is trained using the training data (the model prior to training constitutes a “base machine learning model” because it is the base model prior to training), and (ii) a second stage wherein second training data comprising road” [sensor] “data ([0044] and [0076] sensor data collected while vehicle is traversing roadway; [0099] discussing distinction between vehicle usage sensor data collection and test bench data collection) and environmental data is collected (FIG. 4 vehicle sensor systems 406 include image, audio, temperature and general “environmental” sensors, [0066]; ([0074] vehicle sensor data sent to computer that per [0076]-[0077] uses the data for training) while a plurality of second” [vehicles] “are tested in actual use conditions ([0025] and [0099] in addition to the test bench sensor data, training data may include previously captured sensor data captured by sensors of the vehicle; [0099] discussing distinction between vehicle usage sensor data collection and test bench data collection; [0060] sensor signature data collected for other vehicles; [0076]-[0077] log data 448 used for training may be collected for other vehicles. Examiner notes that collecting the sensor data from other vehicles constitutes testing) until the prominent failure mode occurs in [the vehicle] ([0099] sensor/log data collected for vehicle usage when vehicle experiences a failure or anomaly; [0025] training data may be labeled to indicate ground truth operating status such as indication of wear, indication of anomaly (the labeling of ground truth indicates noted occurrence of the label condition such as wear and/or anomaly)) and wherein the trained base machine learning model is trained using the second training data to create the trained machine learning system ([0101] sensor data used to train model; [0106] model training may be continued (model retrained) using additional sensor data (i.e., previously trained model trained using second training data)), wherein the controller is structured and configured to:
receive the data (FIG. 4 vehicle computing device 404 configured to receive sensor data 436 from sensors 406 and computing device 440 configured to receive log data 448 that per [0076] comprises sensor data received from the vehicle; FIG. 7 block 702, [0099]);
predict wear and tear experienced by the mobility system (Abstract and [0015] disclosing that the model is trained to identify operating status based on sensor input data; [0011] “operating status” may be component wear and/or likelihood of failure). As interpreted in view of Applicant’s specification, the model “identifying” is entailed within a model “prediction” in terms of identifying what the wear/failure condition is expected to be based on the modeling; FIG. 7 step 708 and 710, [0104]) and characterize a lifecycle stage of the” [vehicle] ([0058] monitoring component 430 processes the sensor data to estimate and/or predict vehicle component operation status including indications of wear such as percent of life used/remaining) using the trained machine learning system based on at least the received data ([0077] machine learning model trained to identify indications of wear/remaining life; [0103]-[0104] machine learning model processes the received sensor data to determine the operation status that may include indication of wear); and
generate and transmit an alert for required maintenance for ([0058] monitoring component may provide an “indication of an anomaly” (i.e., indication/alert for needed maintenance); [0091]-[0092] notification may be sent regarding operating status such as wear and remaining life (inherently indicated need for maintenance)) and/or predicted breakdown of the” [vehicle] ([0058] monitoring component generates an operating status that includes an indication of wear (alert related to wear) such as time-to-failure of vehicle; [0091]-[0092] notification may be sent regarding operating status such as wear and remaining life (inherently indicates a failure prediction)) based on the characterized lifecycle stage ([0058] the “indication of wear” and or “indication of anomaly” is/are generated based on the underlying component operation status determination/estimation process; [0091]-[0092] notification based on operating status such as wear and remaining life).”
As set forth above, Claessens teaches using multiple training phases for training a machine learning model and teaches that each of the training phases may entail collecting and using sensor data from a test bench for training and/or using actual use sensor data from multiple vehicles for training the machine learning model. However, Claessens is silent regarding the sequencing of training phases in terms of combining distinct training phases and therefore does not appear to expressly teach that a machine learning model may be trained in two phases in a particular sequence in which the first phase training uses the test bench data and the subsequent training uses the actual use sensor data.
It would have been obvious to one of ordinary skill in the art before the effective filing date, in view of Claessens teaching that either or both test bench and actual usage data training may be effectively used for training the model and further teaching that the model may be optimized by retraining, to have configured Claessens system such that a test bench training phase precedes an actual usage training phase or vice versa because the training options are specified in a limited manner (test bench and/or actual use data may be used for training) such that potential sequences are limited in scope and would be readily appreciated by one of ordinary skill in the art particularly in view of Claessens teaching that either type of training is a valid option for training or retraining.
Regarding the monitored vehicle being a “mobility system,” and the training including collecting training data from a plurality of “mobility systems,” Claessens teaches a system for monitoring a condition of a vehicle defined broadly ([0049]) but does not explicitly teach that the vehicle is an ambulatory device.
Turturiello discloses a sensor based monitoring system (FIG. 1 depicting wheelchair 10 equipped with one or more sensors 110) in which sensor-based condition monitoring is implemented for a wheelchair (mobility device) ([0017]-[0018] operating conditions such as inertial forces tracked by sensors 11 mounted to wheelchair 10).
It would have been obvious to one of ordinary skill in the art before the effective filing date, in view of Turturiello’s teaching of monitoring operating conditions of a mobility device (wheelchair) using multiple sensor for a mobility device (wheelchair) type vehicle and in view of Claessens system for monitoring a condition of a vehicle that in accordance with Claessens is applicable to an open-ended list of vehicle types ([0049] applicable to “any type of vehicle”), to have implemented Claessens system with a wheelchair or other similar mobility device as the vehicle to be monitored by the system components set forth in claim 1, such that the target vehicle is a mobility system with respect to which the model training is accordingly performed by collecting training data for a plurality of other mobility systems.
The motivation would have been to apply Claessens system to a vehicle class that is readily subject to sensor-based condition monitoring as taught by Turturiello such that lifecycle information may be derived in accordance with Claessens disclosed system.
While as set forth above, Claessens teaches that the training data may include kinematic data such as vibration data, neither Claessens nor Turturiello appears to teach collecting and using shock data for training and therefore do not teach the first training data comprising “shock” data and the second training data comprising road “shock” data.
Mhatre discloses a method for evaluating wheelchair castors using “community data” (sensor data obtained from actual usage of multiple wheelchairs) and “established wheelchair test methods” (mechanical test bed) (Abstract; FIG. 2) in which shock data is collected from test bed testing and community testing as particularly relevant to monitoring wheelchair operating condition (pages 4-5, Determining shock exposure based on measurements in the community, paragraph beginning with “Shock exposure is caused by abrupt loads …” through paragraph beginning with “Accelerations on the same castor models …”).
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have applied Mhatre’s teaching of collecting and analyzing shock data from a test bed and from multiple actual usages as particularly significant for assessing the condition of wheelchair type mobility systems to the system taught by Claessens as modified by Turturiello, in which a wheelchair type mobility system is being monitoring using machine learning that uses sensor data experienced by the vehicle to determine vehicle condition, such that in combination the system is configured to collect shock data in both training phases.
The motivation would have been to apply sensor data that as suggested by Mhatre is particularly relevant to assessing wheelchair condition for training the model to accurately assess wheelchair operating status.
As to claim 3, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 1, wherein the lifecycle stage includes a degree of wear on the mobility system (Claessens: [0011] and [0058] operating state may include an indication of wear such as percentage of life used).”
As to claim 4, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 1, wherein the number of sensors includes at least one accelerometer (Claessens: [0011] and [0035] sensors may include an accelerometer).”
As to claim 5, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 1, wherein the number of sensors includes a magnetometer (Claessens: [0066] and [0071] sensors may include a magnetometer).”
As to claim 6, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 1, wherein the number of sensors includes a gyroscope (Claessens: [0066] and [0071] sensors may include a gyroscope).”
As to claim 7, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 1, wherein the number of sensors includes at least one accelerometer and at least one of a magnetometer or a gyroscope (Claessens: [0066] and [0071] sensors may include an accelerometer, a magnetometer, and a gyroscope. Examiner notes that in context of the description of the sensors such as in [0066] and [0071] Claessens indicates that the set may be a combination of these sensors for sensing various difference metrics relating to vehicle usage).”
As to claim 8, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 7, wherein the number of sensors includes at least one accelerometer, a magnetometer, and a gyroscope (Claessens: [0066] and [0071] sensors may include an accelerometer, a magnetometer, and a gyroscope. Examiner notes that in context of the description of the sensors such as in [0066] and [0071] Claessens indicates that the set may be a combination of these sensors for sensing various difference metrics relating to vehicle usage).”
As to claim 9, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 3, further comprising a number of environmental conditions sensors coupled to the mobility system (Claessens: FIG. 4 sensors 406 within vehicle 402 include environmental and temperature sensors), the number of environmental conditions sensors being structured and configured to generate environmental condition data indicative of one or more environmental conditions around the mobility system during use of the mobility system (Claessens: FIG. 4 environmental sensor and temperature sensor among sensors 406 configured to generate environmental and temperature (also environmental) data by definition and being within vehicle 402 are configured to generate environmental data indicative of environmental conditions around the vehicle during use (e.g., temperature sensed by temperature sensor would be influenced by temperature “around” (within or external) the vehicle during use), wherein the trained machine learning system is structured and configured to receive the environmental condition data (Claessens: FIG. 4 monitoring component 430 as part of computing device 404 configured to receive environmental and temperature data from sensors 406, [0058]; FIG. 4 machine learning component 450 configured to receive environmental and temperature data from sensors 406 (received via network 438 and stored as historical sensor data within log data 448), [0062]) and characterize the degree of wear based on the received shock and/or vibration data and the environmental condition data (Claessens: [0015] machine learning model may be trained to detect changes in sensor signals and determine an operating status of a component; [0012] sensor signature may include combinations of sensor data including vibration (acoustic and data captured by IMU that per [0041] indicates vibration) and temperature data; [0058] monitoring component 430 that may be implemented by machine learning determines operating status that may indicate degree of wear such as remaining life).”
As to claim 10, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 9, wherein the number of environmental conditions sensors comprises at least one of a temperature sensor and a humidity sensor (Claessens: FIG. 4 sensors 406 include temperature sensor and other “environmental” sensor; [0035] sensors 102 may include a temperature sensor and a humidity sensor).”
As to claim 11, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 10, wherein the number of environmental condition sensors comprises a temperature sensor and a humidity sensor (Claessens: FIG. 4 sensors 406 include temperature sensor and other “environmental” sensor; [0035] sensors 102 may include a temperature sensor and a humidity sensor).”
As to claim 12, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 3, wherein the trained machine learning system is structured and configured to characterize the degree of wear by determining a usage index (Claessens: FIG. 4 sensor signatures 434 within computing device 404 and sensor signatures 454 within computing device 440; [0015] machine learning model used to generate and determine changes to sensor signature and determine operating status of vehicle; [0013] sensor signature comprises data representative of measurements over time (usage index); [0085] “sensor signature” is indicative of operating status that per [0058] may indicate usage wear; [0058] monitoring component 430 that per [0063] may be implemented as machine learning model characterizes an operating status/indication of wear (sensor signature) in terms of a percentage of life used and/or remaining life) based on at least the received shock and/or vibration data (Claessens: [0058] indication of wear determined by monitoring component 430 based on sensor data from one or more sensors 406 that per FIG. 4 include an inertial sensor that per [0041] measure vibration; [0015] sensor signatures may be based on IMU data), and wherein the controller is structured and configured to generate the alert (Claessens: FIG. 5 block 518, [0092] performing second action may including sending the operating status to a remote computing system) when the determined usage index exceeds a certain predetermined threshold value (Claessens: FIG. 5 blocks 514 and 518, [0090] if difference between sensor signatures indicates substantial variation in similarity (which conveys an association in terms of change of operating status over time as determined by a second signature obtained subsequent in time to the first signature) is outside of some range (variation exceeds a threshold) operation 518 is performed that per [0092] may include sending the operating status to a remote computing system).
As to claim 13, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 12,” and further teaches that Miner’s rule may be used for damage modeling of data over time (wear) to evaluate stress-strain relationships (Claessens: [0024]).
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have applied Claessens teaching of using Miner’s rule for damage wear evaluation to the usage index generation via modeling taught by Claessens such that in combination Miner’s rule is used in the process of generating sensor signatures (usage index) is generated using Miner’s rule.
Such a combination would amount to selecting a known design option for evaluating factors (e.g., stress-strain relationships) that may occur over time to be used damage models to achieve predictable results.
As to claim 14, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 1, wherein the controller is part of the mobility system (Claessens: FIG. 4 vehicle computing device 404 within vehicle 402).”
As to claim 15, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 1, wherein the controller is part of a remote computing system in electrical communication with the mobility system (Claessens: FIG. 4 computing device 440 communicatively coupled (inherently entails electricity/electronics) with vehicle computing device 404).”
As to claim 16, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 1, wherein the trained machine learning system comprises a regression model (Claessens: [0026] and [0065] machine learning model may be a regression model).”
As to claim 18, Claessens teaches “[a] method of monitoring a condition of a” [vehicle] (method performed by FIG. 4 system 400 including sensors 406 and monitoring component 430 configured to monitor vehicle 402; [0011] systems for monitoring vehicle health), comprising:
generating data indicative of use of the” [vehicle] (sensors 102 in FIG. 1 and sensors 406 incorporated with a vehicle and therefore structured and configured to monitor conditions related to use of the vehicle; [0035]-[0036] sensors including IMU, accelerometers, GPS, electrical parameter sensors configured to generate data indicative of operations/use of the vehicle; [0038], [0041], and [0046] sensor generated data may include dynamic kinematic data such as vibration) using a number of sensors (FIG. 1 sensors 102 coupled to vehicle 100; FIG. 2 sensors 406 integrated with vehicle 402), wherein the data includes shock and/or vibration data ([0012] sensor data may include acoustic and/or inertial measurements; [0038] sensor data may include vibration) indicative of one or more shocks and/or vibrations experienced by the" [vehicle] “during use ([0038] and [0041] sensed vibration the result of vibration experienced by vehicle (detection of kinematic metric such as vibration would occur during using vehicle internal or external motion));
receiving the data (FIG. 4 vehicle computing device 404 configured to receive sensor data 436 from sensors 406 and computing device 440 configured to receive log data 448 that per [0076] comprises sensor data received from the vehicle; FIG. 7 block 702, [0099]) in a trained machine learning system (FIG. 4 vehicle computing device 404 implements monitoring component 430 that per [0063] may be implemented as a machine learning model that has been trained; computing device 440 implements machine learning component 450 and training component 446; [0077]), wherein the trained machine learning system has been previously trained ([0063] machine learning model has been trained)” in two stages ([0106] model training may be continued (model retrained) using additional sensor data) “to predict wear and tear experienced by the mobility system during use and failure conditions based on the data indicative of use of the mobility system (Abstract and [0015] disclosing that the model is trained to identify operating status based on sensor input data; [0011] “operating status” may be component wear and/or likelihood of failure). As interpreted in view of Applicant’s specification, the model “identifying” is entailed within a model “prediction” in terms of identifying what the wear/failure condition is expected to be based on the modeling; FIG. 7 step 708 and 710, [0104]), the two stages including: (i) a first stage wherein first training data comprising” [sensor] “data is collected while a first mobility system is tested on a mechanical test bed ([0099] sensor data may be test data for the vehicle be generated on a test bench; [0101] sensor data may be used as training data) until a prominent failure mode occurs ([0025] training data may be labeled to indicate ground truth operating status such as indication of wear, indication of anomaly (the labeling of ground truth indicates noted occurrence of the label condition such as wear and/or anomaly)) and wherein a base machine learning model is trained using the training data (the model prior to training constitutes a “base machine learning model” because it is the base model prior to training), and (ii) a second stage wherein second training data comprising road” [sensor] “data ([0044] and [0076] sensor data collected while vehicle is traversing roadway; [0099] discussing distinction between vehicle usage sensor data collection and test bench data collection) and environmental data is collected (FIG. 4 vehicle sensor systems 406 include image, audio, temperature and general “environmental” sensors, [0066]; ([0074] vehicle sensor data sent to computer that per [0076]-[0077] uses the data for training) while a plurality of second” [vehicles] “are tested in actual use conditions ([0025] and [0099] in addition to the test bench sensor data, training data may include previously captured sensor data captured by sensors of the vehicle; [0099] discussing distinction between vehicle usage sensor data collection and test bench data collection; [0060] sensor signature data collected for other vehicles; [0076]-[0077] log data 448 used for training may be collected for other vehicles. Examiner notes that collecting the sensor data from other vehicles constitutes testing) until the prominent failure mode occurs in [the vehicle] ([0099] sensor/log data collected for vehicle usage when vehicle experiences a failure or anomaly; [0025] training data may be labeled to indicate ground truth operating status such as indication of wear, indication of anomaly (the labeling of ground truth indicates noted occurrence of the label condition such as wear and/or anomaly)) and wherein the trained base machine learning model is trained using the second training data to create the trained machine learning system ([0101] sensor data used to train model; [0106] model training may be continued (model retrained) using additional sensor data (i.e., previously trained model trained using second training data));
predicting wear and tear experienced by the mobility system (Abstract and [0015] disclosing that the model is trained to identify operating status based on sensor input data; [0011] “operating status” may be component wear and/or likelihood of failure). As interpreted in view of Applicant’s specification, the model “identifying” is entailed within a model “prediction” in terms of identifying what the wear/failure condition is expected to be based on the modeling; FIG. 7 step 708 and 710, [0104]) and characterizing a lifecycle stage of the” [vehicle] ([0058] monitoring component 430 processes the sensor data to estimate and/or predict vehicle component operation status including indications of wear such as percent of life used/remaining) using the trained machine learning system based on at least the received data ([0077] machine learning model trained to identify indications of wear/remaining life; [0103]-[0104] machine learning model processes the received sensor data to determining the operation status that may include indication of wear); and
generating and transmitting an alert for required maintenance for ([0058] monitoring component may provide an “indication of an anomaly” (i.e., indication/alert for needed maintenance); [0091]-[0092] notification may be sent regarding operating status such as wear and remaining life (inherently indicated need for maintenance)) and/or predicted breakdown of the” [vehicle] ([0058] monitoring component generates an operating status that includes an indication of wear (alert related to wear) such as time-to-failure of vehicle; [0091]-[0092] notification may be sent regarding operating status such as wear and remaining life (inherently indicates a failure prediction)) based on the characterized lifecycle stage ([0058] the “indication of wear” and or “indication of anomaly” is/are generated based on the underlying component operation status determination/estimation process).”
As set forth above, Claessens teaches using multiple training phases for training a machine learning model and teaches that each of the training phases may entail collecting and using sensor data from a test bench for training and/or using actual use sensor data from multiple vehicles for training the machine learning model. However, Claessens is silent regarding the sequencing of training phases in terms of combining distinct training phases and therefore does not appear to expressly teach that a machine learning model may be trained in two phases in a particular sequence in which the first phase training uses the test bench data and the subsequent training uses the actual use sensor data.
It would have been obvious to one of ordinary skill in the art before the effective filing date, in view of Claessens teaching that either or both test bench and actual usage data training may be effectively used for training the model and further teaching that the model may be optimized by retraining, to have configured Claessens system such that a test bench training phase precedes an actual usage training phase or vice versa because the training options are specified in a limited manner (test bench and/or actual use data may be used for training) such that potential sequences are limited in scope and would be readily appreciated by one of ordinary skill in the art particularly in view of Claessens teaching that either type of training is a valid option for training or retraining.
Regarding the monitored vehicle being a “mobility system,” and the training including collecting training data from a plurality of “mobility systems,” Claessens teaches a method for monitoring a condition of a vehicle defined broadly ([0049]) but does not explicitly teach that the vehicle is an ambulatory device.
Turturiello discloses a sensor based monitoring system/method (FIG. 1 depicting wheelchair 10 equipped with one or more sensors 110) in which sensor-based condition monitoring is implemented for a wheelchair (mobility device) ([0017]-[0018] operating conditions such as inertial forces tracked by sensors 11 mounted to wheelchair 10).
It would have been obvious to one of ordinary skill in the art before the effective filing date, in view of Turturiello’s teaching of monitoring operating conditions of a mobility device (wheelchair) using multiple sensor for a mobility device (wheelchair) type vehicle and in view of Claessens method for monitoring a condition of a vehicle that in accordance with Claessens is applicable to an open-ended list of vehicle types ([0049] applicable to “any type of vehicle”), to have implemented Claessens method with a wheelchair or other similar mobility device as the vehicle to be monitored by the system components set forth in claim 1, such that the target vehicle is a mobility system with respect to which the model training is accordingly performed by collecting training data for a plurality of other mobility systems.
The motivation would have been to apply Claessens system to a vehicle class that is readily subject to sensor-based condition monitoring as taught by Turturiello such that lifecycle information may be derived in accordance with Claessens disclosed system.
While as set forth above, Claessens teaches that the training data may include kinematic data such as vibration data, neither Claessens nor Turturiello appears to teach collecting and using shock data for training and therefore do not teach the first training data comprising “shock” data and the second training data comprising road “shock” data.
Mhatre discloses a method for evaluating wheelchair castors using “community data” (sensor data obtained from actual usage of multiple wheelchairs) and “established wheelchair test methods” (mechanical test bed) (Abstract; FIG. 2) in which shock data is collected from test bed testing and community testing as particularly relevant to monitoring wheelchair operating condition (pages 4-5, Determining shock exposure based on measurements in the community, paragraph beginning with “Shock exposure is caused by abrupt loads …” through paragraph beginning with “Accelerations on the same castor models …”).
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have applied Mhatre’s teaching of collecting and analyzing shock data from a test bed and from multiple actual usages as particularly significant for assessing the condition of wheelchair type mobility systems to the method taught by Claessens as modified by Turturiello, in which a wheelchair type mobility system is being monitoring using machine learning that uses sensor data experienced by the vehicle to determine vehicle condition, such that in combination the system is configured to collect shock data in both training phases.
The motivation would have been to apply sensor data that as suggested by Mhatre is particularly relevant to assessing wheelchair condition for training the model to accurately assess wheelchair operating status.
As to claim 20, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 18, wherein the lifecycle stage includes a degree of wear on the mobility system (Claessens: [0011] and [0058] operating state may include an indication of wear such as percentage of life used).”
As to claim 21, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 18, wherein the number of sensors includes at least one accelerometer (Claessens: [0011] and [0035] sensors may include an accelerometer).”
As to claim 22, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 18, wherein the number of sensors includes a magnetometer (Claessens: [0066] and [0071] sensors may include a magnetometer).”
As to claim 23, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 18, wherein the number of sensors includes a gyroscope (Claessens: [0066] and [0071] sensors may include a gyroscope).”
As to claim 24, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 18, wherein the number of sensors includes at least one accelerometer and at least one of a magnetometer or a gyroscope (Claessens: [0066] and [0071] sensors may include an accelerometer, a magnetometer, and a gyroscope. Examiner notes that in context of the description of the sensors such as in [0066] and [0071] Claessens indicates that the set may be a combination of these sensors for sensing various difference metrics relating to vehicle usage).”
As to claim 25, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 24, wherein the number of sensors includes at least one accelerometer, a magnetometer, and a gyroscope (Claessens: [0066] and [0071] sensors may include an accelerometer, a magnetometer, and a gyroscope. Examiner notes that in context of the description of the sensors such as in [0066] and [0071] Claessens indicates that the set may be a combination of these sensors for sensing various difference metrics relating to vehicle usage).”
As to claim 26, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 18, further comprising generating environmental condition data indicative of one or more environmental conditions around the mobility system during use of the mobility system (Claessens: FIG. 4 environmental sensor and temperature sensor among sensors 406 configured to generate environmental and temperature (also environmental) data by definition and being within vehicle 402 are configured to generate environmental data indicative of environmental conditions around the vehicle during use (e.g., temperature sensed by temperature sensor would be influenced by temperature “around” (within or external) the vehicle during use) using a number of environmental conditions sensors coupled to the mobility system (Claessens: FIG. 4 sensors 406 within vehicle 402 include environmental and temperature sensors), and receiving the environmental condition data in the trained machine learning system (Claessens: FIG. 4 monitoring component 430 as part of computing device 404 configured to receive environmental and temperature data from sensors 406, [0058]; FIG. 4 machine learning component 450 configured to receive environmental and temperature data from sensors 406 (received via network 438 and stored as historical sensor data within log data 448), [0062]), wherein the characterizing comprises characterizing the degree of wear based on the received shock and/or vibration data and the environmental condition data (Claessens: [0015] machine learning model may be trained to detect changes in sensor signals and determine an operating status of a component; [0012] sensor signature may include combinations of sensor data including vibration (acoustic and data captured by IMU that per [0041] indicates vibration) and temperature data; [0058] monitoring component 430 that may be implemented by machine learning determines operating status that may indicate degree of wear such as remaining life).”
As to claim 27, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 26, wherein the number of environmental conditions sensors comprises at least one of a temperature sensor and a humidity sensor (Claessens: FIG. 4 sensors 406 include temperature sensor and other “environmental” sensor; [0035] sensors 102 may include a temperature sensor and a humidity sensor).”
As to claim 28, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 27, wherein the number of environmental condition sensors comprises a temperature sensor and a humidity sensor (Claessens: FIG. 4 sensors 406 include temperature sensor and other “environmental” sensor; [0035] sensors 102 may include a temperature sensor and a humidity sensor).”
As to claim 29, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 20, wherein the trained machine learning system is structured and configured to characterize the degree of wear by determining a usage index (Claessens: FIG. 4 sensor signatures 434 within computing device 404 and sensor signatures 454 within computing device 440; [0015] machine learning model used to generate and determine changes to sensor signature and determine operating status of vehicle; [0013] sensor signature comprises data representative of measurements over time (usage index); [0085] “sensor signature” is indicative of operating status that per [0058] may indicate usage wear; [0058] monitoring component 430 that per [0063] may be implemented as machine learning model characterizes an operating status/indication of wear (sensor signature) in terms of a percentage of life used and/or remaining life) based on at least the received shock and/or vibration data (Claessens: [0058] indication of wear determined by monitoring component 430 based on sensor data from one or more sensors 406 that per FIG. 4 include an inertial sensor that per [0041] measure vibration; [0015] sensor signatures may be based on IMU data), and wherein the alert is generated (Claessens: FIG. 5 block 518, [0092] performing second action may including sending the operating status to a remote computing system) when the determined usage index exceeds a certain predetermined threshold value (Claessens: FIG. 5 blocks 514 and 518, [0090] if difference between sensor signatures indicates substantial variation in similarity (which conveys an association in terms of change of operating status over time as determined by a second signature obtained subsequent in time to the first signature) is outside of some range (variation exceeds a threshold) operation 518 is performed that per [0092] may include sending the operating status to a remote computing system).”
As to claim 30, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 29,” and further teaches that Miner’s rule may be used for damage modeling of data over time (wear) to evaluate stress-strain relationships (Claessens: [0024]).
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have applied Claessens teaching of using Miner’s rule for damage wear evaluation to the usage index generation via modeling taught by Claessens such that in combination Miner’s rule is used in the process of generating sensor signatures (usage index) is generated using Miner’s rule.
Such a combination would amount to selecting a known design option for evaluating factors (e.g., stress-strain relationships) that may occur over time to be used damage models to achieve predictable results.
As to claim 31, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 18, wherein the trained machine learning system is provided as part of the mobility system (Claessens: FIG. 4 vehicle computing device 404 within vehicle 402).”
As to claim 32, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 18, wherein the trained machine learning system is provided as part of a remote computing system in electrical communication with the mobility system (Claessens: FIG. 4 computing device 440 communicatively coupled (inherently entails electricity/electronics) with vehicle computing device 404).”
As to claim 33, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 18, wherein the trained machine learning system comprises a regression model (Claessens: [0026] and [0065] machine learning model may be a regression model).”
As to claim 35, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 1, wherein the alert is electronically transmitted to a user computing device (Claessens: [0091]-[0092] notification may be sent to a computer system. Examiner notes that a broadest reasonable interpretation of “user computing device” may include any computer that is utilized for or by a person, which encompasses practically any type of computer system whose function relates to persons such as the “remote computing system” in [0092] that is associated with autonomous vehicle fleets that inherently relate to passengers) and/or a cloud-based registry for mobility system users configured to send a notification to one or more third parties in response to the alert.”
Even if “user computing device” is interpreted more narrowly to encompass a computer device directly used by a user for obtaining the alert, Claessens discloses that it is generally known to provide automated vehicle condition information to a user ([0001] and [0010] vehicle may notify user of vehicle condition including failure condition).
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have combined Claessens teaching of providing direct, automated notification of vehicle condition to users with Claessens teaching of vehicular computerized system configured to provide the notice (as in [0001] and FIG. 4 vehicle computing device 404) and teaching that the alert may be sent to a computer such that the system is configured to electronically transmit the alert to a “user” computing device.
The motivation would have been to provide a vehicle user/owner notice of a vehicle condition status that may require use attention as suggested by Claessens.
As to claim 36, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 1” [claim 18], “wherein the alert is electronically transmitted to a user computing device (Claessens: [0091]-[0092] notification may be sent to a computer system. Examiner notes that a broadest reasonable interpretation of “user computing device” may include any computer that is utilized for or by a person, which encompasses practically any type of computer system whose function relates to persons such as the “remote computing system” in [0092] that is associated with autonomous vehicle fleets that inherently relate to passengers) and/or a cloud-based registry for mobility system users configured to send a notification to one or more third parties in response to the alert.
Even if “user computing device” is interpreted more narrowly to encompass a computer device directly used by a user for obtaining the alert, Claessens discloses that it is generally known to provide automated vehicle condition information to a user ([0001] and [0010] vehicle may notify user of vehicle condition including failure condition).
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have combined Claessens teaching of providing direct, automated notification of vehicle condition to users with Claessens teaching of vehicular computerized system configured to provide the notice (as in [0001] and FIG. 4 vehicle computing device 404) and teaching that the alert may be sent to a computer such that the system is configured to electronically transmit the alert to a “user” computing device.
The motivation would have been to provide a vehicle user/owner notice of a vehicle condition status that may require use attention as suggested by Claessens.
Claims 17 and 34 are rejected under 35 U.S.C. 103 as being unpatentable over Claessens in view of Turturiello and Mhatre as applied to claims 16 and 33 above, and further in view of Chen (US 2017/0314961 A1).
As to claim 17 the combination of Claessens, Turturiello, and Mhatre teaches “[t]he system according to claim 16,” and further teaches that a variety of different types of models may be used (Claessens: [0026] and [0065]) but does not expressly teach “wherein the regression model is a time-series regression model.”
Chen discloses a system for anomaly detection over time for physical systems (Abstract; FIG. 1) that includes using a time-series regression model for monitoring temporal evaluation of the condition of a system ([0023] modeling engine may include AutoRegressive model with eXternal input (ARX) for analyzing time series data that follows a property; [0061] Autoregressive model used to measure linear signal dynamics; [0079] ARX model used to capture dependencies of current values of two time series (determines/predicts future responses based on response history and transfer of dynamics from relevant predictors consistent with the known function of AutoRegressive model with eXternal input (ARX))).
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have applied Chen’s teaching of using a time-series regression model for monitoring the functioning of a system to the system taught by Claessens as modified by Turturiello and Mhatre such that in combination the regression model is a time-series regression model.
The motivation would have been to evaluate temporal dynamics in the sensor data, which as disclosed by each of Claessens and Chen are monitored over time, to be used for more effective condition monitoring provided by the modeling as suggested by Chen.
As to claim 34, the combination of Claessens, Turturiello, and Mhatre teaches “[t]he method according to claim 33,” and further teaches that a variety of different types of models may be used (Claessens: [0026] and [0065]) but does not expressly teach “wherein the regression model is a time-series regression model.”
Chen discloses a system for anomaly detection over time for physical systems (Abstract; FIG. 1) that includes using a time-series regression model for monitoring temporal evaluation of the condition of a system ([0023] modeling engine may include AutoRegressive model with eXternal input (ARX) for analyzing time series data that follows a property; [0061] Autoregressive model used to measure linear signal dynamics; [0079] ARX model used to capture dependencies of current values of two time series (determines/predicts future responses based on response history and transfer of dynamics from relevant predictors consistent with the known function of AutoRegressive model with eXternal input (ARX))).
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have applied Chen’s teaching of using a time-series regression model for monitoring the functioning of a system to the method taught by Claessens as modified by Turturiello and Mhatre such that in combination the regression model is a time-series regression model.
The motivation would have been to evaluate temporal dynamics in the sensor data, which as disclosed by each of Claessens and Chen are monitored over time, to be used for more effective condition monitoring provided by the modeling as suggested by Chen.
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 MATTHEW W BACA whose telephone number is (571)272-2507. The examiner can normally be reached Monday - Friday 8:00 am - 5:30 pm.
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, Andrew Schechter can be reached at (571) 272-2302. 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.
/MATTHEW W. BACA/Examiner, Art Unit 2857
/ANDREW SCHECHTER/Supervisory Patent Examiner, Art Unit 2857