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 Amendments
The amendment filed 10/29/2025 has been entered.
Claims 1-20 remain pending in the application.
Claims 1, 8, and 15 have been amended.
The amendment filed 10/29/2025 is sufficient to overcome the 101 rejections of
claims 1-20. The previous rejections have been withdrawn.
The amendment filed 10/29/2025 is sufficient to overcome the 35 U.S.C. 103 rejections of claims 1-5, 7-12, and 14-20 over Yamamoto in view of Abbott and Alam and the 35 U.S.C. 103 rejections of claims 6 and 13 over Yamamoto in view of Abbott and Alam and further in view of Mahapatra. The previous rejections have been withdrawn.
Argument 1, regarding the 112(b) rejections, applicant argues that the 112(b) rejection of claim 18 should be withdrawn in view of amendments. Examiner agrees and the 112(b) rejection of claim 18 has been withdrawn.
Argument 2, regarding the 101 rejections, applicant argues that the claims integrate the abstract ideas into a practical application of effective battery management to charge a vehicle battery in accordance to a charge command output by a model. Examiner agrees and the 101 rejections have been withdrawn.
Argument 3, regarding the prior art rejections, Applicant argues that Abbot does not teach approximating hidden battery state information and instead teaches collecting sensor data to infer external battery state information. Examiner notes this argument is moot in view of Zhang et al (US 20200175212 A1). Zhang teaches wherein the hidden battery state information comprises unobservable internal variables of the battery that are not directly measured by sensors (battery models may be directed towards mechanisms such as internal characteristics of batteries. Electrochemical models describe internal chemical processes of batteries by using complex nonlinear differential equations, and models may include a Markov process, P0003);…and controlling a battery management system to charge the vehicle battery in accordance with the charge command output by the actor model (“Battery models are of great significance for the rational design and safe operation of power batteries and battery management systems thereof, and are the basis of battery state of charge (SOC) estimation, state of health (SOH) estimation, remaining useful life (RUL) prediction, etc.”, P0003. Batteries may be electric vehicle batteries, P0002).
Accordingly, it would have been obvious to a person having ordinary skill in the
art before the effective filing date of the claimed invention, having the teachings of Yamamoto, Abbott, Alam, and Zhang before them, to include Zhang’s specific teaching of electrochemical models being used for battery management in Yamamoto’s system of generating a machine learning model for an electric vehicle. One would have been motivated to make such a combination of using an electrochemical model to describe internal chemical processes of a battery (see Zhang P0002-P0003) and using a recurrent neural network to train machine learning model 1M (see Yamamoto P0127).
Applicant also argues that the Yamamoto reference is not relevant to the current invention because it is directed towards the technical problem of battery discharge management for user comfort in e-bikes. Examiner respectfully disagrees because under the broadest reasonable interpretation, an e-bike is still interpreted as a vehicle and the current invention is directed towards effective battery management of a vehicle battery, which is reflected in Yamamoto.
Applicant also argues that the Yamamoto reference does not properly teach “wherein a loss of the critic is backpropagated through the critic and through the SPNNFE” as recited in claims 4, 11, and 17 because the reference does not explicitly mention backpropagating loss but instead merely recites training the RNN with the use of backpropagation. Examiner respectfully disagrees because although the reference does not explicitly recite “loss” or “error” being backpropagated, the act of backpropagation is defined as calculating how much each network weight contributes to the overall loss or error and using that information to adjust weights backward through layers to improve accuracy. Thus, the teaching of backpropagation includes the teaching of backpropagating loss through the recurrent neural network. Regarding dependent claims 5, 12, and 18, applicant argues that the cited art does not teach freezing weights during backpropagation. Examiner notes that the claims themselves do not recite this limitation, but instead recite not updating weights during backpropagation which currently reads as a negative limitation.
The full prior art rejections are outlined below.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph.
A broad range or limitation together with a narrow range or limitation that falls within the broad range or limitation (in the same claim) may be considered indefinite if the resulting claim does not clearly set forth the metes and bounds of the patent protection desired. See MPEP § 2173.05(c). In the present instance, claims 1, 8, and 15 recite the broad recitation “via one or more electronic battery sensors, determining… hidden battery state information”, and the claim also recites “wherein the hidden battery state information comprises unobservable internal variables of the battery that are not directly measured by sensors” which is the narrower statement of the range/limitation. The claim(s) are considered indefinite because there is a question or doubt as to whether the feature introduced by such narrower language is (a) merely exemplary of the remainder of the claim, and therefore not required, or (b) a required feature of the claims.
Dependent claims 2-7, 9-14, and 16-20 depend on claims 1, 8, and 15 and inherit the deficiency.
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 1-5, 7-12, and 14-20 are rejected under 35 U.S.C. 103 as being unpatentable over Yamamoto et al (Pub. No.: US 20210009226 A1), hereafter Yamamoto in view of Abbott et al (Pub. No.: US 20240429730 A1), hereafter Abbott, Alam et al (Pub. No.: US 20230206111 A1), hereafter Alam, and Zhang et al (US 20200175212 A1), hereafter Zhang.
Regarding claims 1 and 8, Yamamoto teaches providing a reinforcement learning model including (i) an actor model configured to produce an output associated with a charge command to charge the vehicle battery, and (ii) a critic model configured to output a predicted reward (Actor-critic reinforcement learning algorithm where the actor model is associated with the action of charging a battery and the critic model outputs a reward based on the electric power consumption amount during a preset period for every preset period until the battery H1 reaches a preset ratio or lower from the full charge, P0126); and training the reinforcement learning model based on (i) the vehicle battery state information, and (ii) the extracted features (Reinforcement learning includes updating weights to improve electric power consumption efficiency in the battery, P0105); wherein the training includes: updating weights of the actor model to maximize the predicted reward output by the critic model, …and updating weights of the critic model to minimize a difference between (i) the predicted reward output by the critic model and (ii) health-based rewards received from charging of the vehicle battery (When implementing reinforcement learning with an actor critic model, weights may be adjusted to improve the index value, P0124-P0126. The index value indicates electric power consumption efficiency of a battery, a measure of a battery’s health, P0104).
Yamamoto does not appear to explicitly teach via one or more electronic battery sensors, determining observable battery state data associated with charging of a vehicle battery, wherein vehicle battery state information includes the observable battery state data and hidden battery state information;… and approximating at least some of the hidden battery state information based on the extracted features in order to optimize charging of the vehicle battery.
Abbott teaches via one or more electronic battery sensors, determining observable battery state data associated with charging of a vehicle battery, wherein vehicle battery state information includes the observable battery state data and hidden battery state information (Sensors collect data associated with battery charging. The sensor data may include observable battery data such as a measured current or temperature of a batter, or hidden data such as accelerometer data, gyroscope data, and/or magnetometer data, P0040, P0043. The battery may be a vehicle battery, P0038);… and approximating at least some of the hidden battery state information based on the extracted features in order to optimize charging of the vehicle battery (“inertial sensor data (e.g., accelerometer data, gyroscope data, magnetometer data) can be used to determine a position of the power tool battery charger 702…These data can be used to generate charger operation data to control the charging action of the charging circuit(s) 758 in an optimized manner for the current usage of the power tool battery charger 702 and/or for future likely usage of the power tool battery charger 702”, P0117-P0118).
Accordingly, it would have been obvious to a person having ordinary skill in the
art before the effective filing date of the claimed invention, having the teachings of Yamamoto and Abbott before them, to include Abbott’s specific teaching of collecting sensor data related to charging vehicle batteries in Yamamoto’s system of generating a machine learning model for an electric vehicle. One would have been motivated to make such a combination of collecting sensor data related to a vehicle’s battery (see Abbott P0038, P0040, P0043) and collecting information with speed, acceleration, torque, and angle sensors to collect information regarding vehicle A (see Yamamoto P0059).
Yamamoto in view of Abbot does not appear to explicitly teach via a sequence-processing neural network feature extractor (SPNNFE), extracting features from preceding vehicle battery state information;…and updating weights of the SPNNFE.
Alam teaches via a sequence-processing neural network feature extractor (SPNNFE), extracting features from preceding vehicle battery state information (In view of P0045 of the specification of the instant application, SPNNFE is a broader RNN feature extractor. RNN may be used to extract features from battery health data, P0007. Prognostic tasks carried out by ML models include processing sequential data, P0002-P0003);…and updating weights of the SPNNFE (During model training, RNN model may be trained or fine-tuned based on the current prognostics task, which may be improving battery health, P0040, P0007).
Accordingly, it would have been obvious to a person having ordinary skill in the
art before the effective filing date of the claimed invention, having the teachings of Yamamoto, Abbott, and Alam before them, to include Alam’s specific teaching of a recurrent neural network feature extractor in Yamamoto’s system of generating a machine learning model for an electric vehicle. One would have been motivated to make such a combination of using a recurrent neural network feature extractor (see Alam P0002-P0003, P0007) and using a recurrent neural network to train machine learning model 1M (see Yamamoto P0127).
Yamamoto in view of Abbot and Alam does not appear to explicitly teach “wherein the hidden battery state information comprises unobservable internal variables of the battery that are not directly measured by sensors… and controlling a battery management system to charge the vehicle battery in accordance with the charge command output by the actor model”.
Zhang teaches wherein the hidden battery state information comprises unobservable internal variables of the battery that are not directly measured by sensors (battery models may be directed towards mechanisms such as internal characteristics of batteries. Electrochemical models describe internal chemical processes of batteries by using complex nonlinear differential equations, and models may include a Markov process, P0003);…and controlling a battery management system to charge the vehicle battery in accordance with the charge command output by the actor model (“Battery models are of great significance for the rational design and safe operation of power batteries and battery management systems thereof, and are the basis of battery state of charge (SOC) estimation, state of health (SOH) estimation, remaining useful life (RUL) prediction, etc.”, P0003. Batteries may be electric vehicle batteries, P0002).
Accordingly, it would have been obvious to a person having ordinary skill in the
art before the effective filing date of the claimed invention, having the teachings of Yamamoto, Abbott, Alam, and Zhang before them, to include Zhang’s specific teaching of electrochemical models being used for battery management in Yamamoto’s system of generating a machine learning model for an electric vehicle. One would have been motivated to make such a combination of using an electrochemical model to describe internal chemical processes of a battery (see Zhang P0002-P0003) and using a recurrent neural network to train machine learning model 1M (see Yamamoto P0127).
Regarding claim 15, Yamamoto teaches providing a reinforcement learning model including (i) an actor model configured to produce an output associated with a control system, and (ii) a critic model configured to output a predicted reward; (Actor-critic reinforcement learning algorithm where the actor model is associated with the action of charging a battery and the critic model outputs a reward based on the electric power consumption amount during a preset period for every preset period until the battery H1 reaches a preset ratio or lower from the full charge, P0126); training the reinforcement learning model based on the state information and the extracted features (Reinforcement learning includes updating weights to improve electric power consumption efficiency in the battery, P0105); wherein the training includes: updating weights of the actor model to maximize the predicted reward output by the critic model, …and updating weights of the critic model to minimize a difference between (i) the predicted reward output by the critic model and (ii) rewards associated with the output for the control system (When implementing reinforcement learning with an actor critic model, weights may be adjusted to improve the index value, P0124-P0126. The index value indicates electric power consumption efficiency of a battery, a measure of a battery’s health, P0104).
Yamamoto does not appear to explicitly teach via one or more electronic battery sensors, determining observable battery state data associated with charging of a vehicle battery, wherein vehicle battery state information includes the observable battery state data and hidden battery state information;… and approximating at least some of the hidden battery state information based on the extracted features in order to optimize charging of the vehicle battery.
Abbott teaches via one or more electronic sensors, determining observable state information, wherein state information includes the observable state information and hidden state information (Sensors collect data associated with battery charging. The sensor data may include observable battery data such as a measured current or temperature of a batter, or hidden data such as accelerometer data, gyroscope data, and/or magnetometer data, P0040, P0043. The battery may be a vehicle battery, P0038);… and using the trained reinforcement learning model to approximate at least some of the hidden state information based on the extracted features (“inertial sensor data (e.g., accelerometer data, gyroscope data, magnetometer data) can be used to determine a position of the power tool battery charger 702…These data can be used to generate charger operation data to control the charging action of the charging circuit(s) 758 in an optimized manner for the current usage of the power tool battery charger 702 and/or for future likely usage of the power tool battery charger 702”, P0117-P0118).
Accordingly, it would have been obvious to a person having ordinary skill in the
art before the effective filing date of the claimed invention, having the teachings of Yamamoto and Abbott before them, to include Abbott’s specific teaching of collecting sensor data related to charging vehicle batteries in Yamamoto’s system of generating a machine learning model for an electric vehicle. One would have been motivated to make such a combination of collecting sensor data related to a vehicle’s battery (see Abbott P0038, P0040, P0043) and collecting information with speed, acceleration, torque, and angle sensors to collect information regarding vehicle A (see Yamamoto P0059).
Yamamoto in view of Abbot does not appear to explicitly teach via a sequence-processing neural network feature extractor (SPNNFE), extracting features from preceding vehicle battery state information;…and updating weights of the SPNNFE.
Alam teaches via a recurrent neural network feature extractor (RRNFE), extracting features from preceding state information (RNN may be used to extract features from battery health data, P0007. Prognostic tasks carried out by ML models include processing sequential data, P0002-P0003);… and updating weights of the RRNFE (During model training, RNN model may be trained or fine-tuned based on the current prognostics task, which may be improving battery health, P0040, P0007).
Accordingly, it would have been obvious to a person having ordinary skill in the
art before the effective filing date of the claimed invention, having the teachings of Yamamoto, Abbott, and Alam before them, to include Alam’s specific teaching of a recurrent neural network feature extractor in Yamamoto’s system of generating a machine learning model for an electric vehicle. One would have been motivated to make such a combination of using a recurrent neural network feature extractor (see Alam P0002-P0003, P0007) and using a recurrent neural network to train machine learning model 1M (see Yamamoto P0127).
Yamamoto in view of Abbot and Alam does not appear to explicitly teach “wherein the hidden battery state information comprises unobservable internal variables of the battery that are not directly measured by sensors… and controlling a battery management system to charge the vehicle battery in accordance with the charge command output by the actor model”.
Zhang teaches wherein the hidden battery state information comprises unobservable internal variables of the battery that are not directly measured by sensors (battery models may be directed towards mechanisms such as internal characteristics of batteries. Electrochemical models describe internal chemical processes of batteries by using complex nonlinear differential equations, and models may include a Markov process, P0003);…and controlling a battery management system to charge the vehicle battery in accordance with the charge command output by the actor model (“Battery models are of great significance for the rational design and safe operation of power batteries and battery management systems thereof, and are the basis of battery state of charge (SOC) estimation, state of health (SOH) estimation, remaining useful life (RUL) prediction, etc.”, P0003. Batteries may be electric vehicle batteries, P0002).
Accordingly, it would have been obvious to a person having ordinary skill in the
art before the effective filing date of the claimed invention, having the teachings of Yamamoto, Abbott, Alam, and Zhang before them, to include Zhang’s specific teaching of electrochemical models being used for battery management in Yamamoto’s system of generating a machine learning model for an electric vehicle. One would have been motivated to make such a combination of using an electrochemical model to describe internal chemical processes of a battery (see Zhang P0002-P0003) and using a recurrent neural network to train machine learning model 1M (see Yamamoto P0127).
Regarding claims 2 and 9, Yamamoto in view of Abbott, Alam, and Zhang teaches the elements of claims 1 and 8 as outlined above. Alam further teaches wherein the SPNNFE includes a recurrent neural network (RNN) (Machine learning model used to process sequential data may be a recurrent neural network, P0007).
Regarding claims 3, 10, and 16, Yamamoto in view of Abbott, Alam, and Zhang teaches the elements of claims 1, 8, and 15 as outlined above. Abbott further teaches wherein the one or more electronic battery sensors includes one or more of a voltage sensor, a current sensor, and a temperature sensor (Sensors collect data associated with battery charging. The sensor data may include observable battery data such as a measured current or temperature of a batter, or hidden data such as accelerometer data, gyroscope data, and/or magnetometer data, P0040. The battery may be a vehicle battery, P0038).
Regarding claims 4 and 11, Yamamoto in view of Abbott, Alam, and Zhang teaches the elements of claims 1 and 8 as outlined above. Yamamoto further teaches wherein a loss of the critic is backpropagated through the critic and through the SPNNFE (Actor-critic model may be trained with backpropagation. A recurrent neural network (RNN) may be the learning method for model 1M, P0126-P0127). Alam further teaches …in order to modify the weights of the SPNNFE (Recurrent neural network used to perform prognostic tasks such as improving battery health may be trained, P0007, P0040).
Regarding claim 17, Yamamoto in view of Abbott, Alam, and Zhang teaches the elements of claim 15 as outlined above. Yamamoto further teaches wherein a loss of the critic is backpropagated through the critic and through the RRNFE (Actor-critic model may be trained with backpropagation. A recurrent neural network (RNN) may be the learning method for model 1M, P0126-P0127). Alam further teaches …in order to modify the weights of the RRNFE (Recurrent neural network used to perform prognostic tasks such as improving battery health may be trained, P0007, P0040).
Regarding claims 5, 12, and 18, Yamamoto in view of Abbott, Alam, and Zhang teaches the elements of claims 4, 11, and 17 as outlined above. Alam further teaches wherein during backpropagation of the actor model, the weights of the SPNNFE are not updated (Training used for recurrent neural network that performs prognostic tasks such as improving battery health does not recite backpropagation, P0007, P0040).
Regarding claims 7 and 14, Yamamoto in view of Abbott, Alam, and Zhang teaches the elements of claims 1 and 8 as outlined above. Abbott further teaches wherein the training of the reinforcement learning model is also based upon a current applied to the battery (Machine learning controller 715 may implement reinforcement learning based on data including a charging current applied to a battery, P0255, P0257).
Regarding claim 19, Yamamoto in view of Abbott, Alam, and Zhang teaches the elements of claim 15 as outlined above. Abbott further teaches wherein the observable state information includes information associated with a state of charge of a vehicle battery (Sensors may collect usage information including the state of charge of the battery, P0043).
Regarding claim 20, Yamamoto in view of Abbott, Alam, and Zhang teaches the elements of claim 19 as outlined above. Yamamoto further teaches wherein the updating of the weights of the actor model is made on a charge cycle-by-cycle basis of the vehicle battery (Model may be trained with reinforcement learning, with rewards being given based on electric power consumption until the battery reaches a preset ratio or lower from the full charge).
Claims 6 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Yamamoto in view of Abbott, Alam, and Zhang and further in view of Mahapatra et al (Pub. No.: US 12045272 B2), hereafter Mahapatra.
Regarding claims 6 and 13, Yamamoto in view of Abbott, Alam, and Zhang teaches the elements of claim 1 and 8 as outlined above. Yamamoto does not appear to explicitly teach outputting a trained reinforcement learning model and a trained SPNNFE based on convergence.
Mahapatra teaches outputting a trained reinforcement learning model and a trained SPNNFE based on convergence (When training a reinforcement learning model and RNN, training continues until convergence, C10:L65-67, C11:L1-14).
Accordingly, it would have been obvious to a person having ordinary skill in the
art before the effective filing date of the claimed invention, having the teachings of Yamamoto, Abbott, Alam, Zhang and Mahapatra before them, to include Mahapatra’s specific teaching of training a recurrent neural network and RNN until convergence in Yamamoto’s system of generating a machine learning model for an electric vehicle. One would have been motivated to make such a combination of training a recurrent neural network and RNN until convergence (see Mahapatra C10:L65-67, C11:L1-14) and using a recurrent neural network and reinforcement learning to train machine learning model 1M (see Yamamoto P0127).
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 ISHAN MOUNDI whose telephone number is (703)756-1547. The examiner can normally be reached 8:30 A.M. - 5 P.M..
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, Matthew Ell can be reached at (571) 270-3264. 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.
/I.M./Examiner, Art Unit 2141
/MATTHEW ELL/Supervisory Patent Examiner, Art Unit 2141