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 .
Response to Arguments
Applicant’s arguments received 27 February 2026, have been fully considered. Claims 1-2, 4-6, and 8-110 are pending. Claims 1, 4, and 6 have been amended.
Applicant’s arguments that the claims should be eligible under 35 U.S.C. 101 have been considered but are not persuasive.
Applicant argues that the claims are not merely directed to a mathematical process; unlike a process which depends on predefined equations or physical assumptions, the claims use actual data to determine interdependencies to infer aging state. Applicant further argues that the aging determination considers complex interdependencies among various data. Applicant argues that machine learning algorithms are dissimilar to applying fixed mathematical formulas, and that the approach is inherently non-mathematical. Finally, Applicant argues that the claims recite a technological advancement in battery aging computation.
The examiner agrees with some of Applicant’s statements but disagrees with Applicant’s conclusions. The examiner maintains that “calculating, by the machine learning architecture, an aging state” recites a mathematical process, as the output would be quantitative and would be obtained by the machine learning “calculating”. Nevertheless, the examiner agrees that mathematical processes are not described in detail.
The examiner disagrees that this should make the claims eligible under 35 U.S.C. 101. The claims in effect recite collecting data known to relate to battery aging, then applying in general terms a machine learning model to determine a battery’s state of health and using that information to “maintain…a range of the battery.” In other words, the claims are little more than applying the concept of machine learning to a calculation (see MPEP 2106.05(f)). While the aging state of a battery may be practical information, using a generic technique to obtain it is equivalent to simply applying machine learning to the general field of battery health (see MPEP 2106.05(h)).
See 101 rejections below.
Applicant’s arguments regarding the prior art rejections under 35 U.S.C. 103 have been considered.
Applicant argues that Simonis does not disclose using historical aging states as inputs, however the examiner disagrees that this is necessary as claim 1 recites (and claims 4 and 6 recite essentially the same): “calculating, by the machine learning architecture, an aging state of the at least one cell of the battery…the aging state calculation using…the historical data related to the aging states of at least one cell of other batteries.” The examiner considers that using a machine learning architecture trained on historical aging data of other batteries would satisfy this, regardless of whether the historical data as used as an input into a trained model. Furthermore, Simonis does describe that a machine learning (ML) model is trained on historical data (¶49: “The state of health model can be continually updated or retrained in a central processing unit based on operating variables of the vehicle batteries from the vehicle fleet.”). Also, it would have been obvious to train a SOH-estimation ML model on historical aging data from other batteries; if a ML model is meant to predict output
y
, it is typically trained on historical data of inputs
x
and outputs
y
so it can learn how
y
depends on
x
.
Applicant’s amendments recite limitations that would be self-evident in the context of the previous rejection. The method would naturally be “computer-implemented”; the ML architecture would be stored in a memory so it can be accessed by a computing system; the ML would be trained to accept a sequence of variables representing driver behavior (thus “learning” the behavior over time) and to estimate its battery’s aging state based on the driver behavior and historical data on which the ML was trained; the ML would perform the calculation; and the calculation would be based on the historical data used to train the ML.
See 103 rejections below.
Claim Objections
Claims objected to because of the following informalities:
In claim 1, step c, “a computing system” should be replaced with “the computing system”.
In claim 6, line 5, “…code including instructions a machine learning algorithm” should be amended to read: “…code including instructions for a machine learning algorithm”.
In claim 6, line 9, the period should be removed in “historical aging states.”
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 1-2, 4-6, and 8-10 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
At Step 1 of the 101 analysis, all claims are directed to one of the statutory categories
Claim 1 is rejected in response to the following analysis:
At Step 2A Prong One, the judicial exceptions are bolded in the copy of claim 1 below:
A computer-implemented method for determining an aging state of at least one cell of a battery, said method comprising the following steps:
storing a machine learning architecture in memory of a computing system;
training the machine learning architecture for learning a driving behavior of a user of a motor vehicle, and for determining the aging state of at least one cell of the battery of the motor vehicle based on the learned driving behavior and historical data related to aging states of at least one cell of other batteries;
receiving, by at least one sensor of the computing system, operating data relating to the at least one cell of the battery, wherein the operating data is assigned a specified timestamp based on a point of detection by the at least one sensor, the operating data including at least: an electrical charge and electrical voltage across the at least one cell of the battery and torque requests to an electric drive motor;
storing the received operating data in a memory unit of the computing system with the specified timestamp;
calculating, by the machine learning architecture, an aging state of the at least one cell of the battery using a processor of the computing system while interdependently taking into account multiple parameters included in the operating data at common timestamps, the aging state calculation using a power output measured at plural common timestamps, a physical condition of the at least one cell of the battery determined from physically measured values, and external conditions determined from the operating data, and the historical data related to the aging states of at least one cell of other batteries; and
maintaining, by the processor of the computing system, a range of the battery using the calculated aging state to model an interdependency of the multiple parameters of the operating data associated with the power output of at least two common timestamps of the plural common timestamps and the physically measured values and adjust one or more of the multiple parameters or factors relevant to the aging state.
Calculating an aging state of one or more battery cells while taking into account data is a mathematical process.
At Step 2A Prong Two, the additional elements do not integrate the judicial exception into a practical application. The additional elements are: a battery comprising at least one cell; a computing system comprising a sensor, a memory unit, and a processor; training a machine learning architecture on driving behavior and historical data; receiving data, assigning a timestamp to it, and storing it in the memory unit; using the processor to calculate the aging state; and maintaining the battery range by adjusting one or more parameters or factors using the processor based on the aging state calculation. Since the battery is rechargeable but not otherwise limited, the battery range may be interpreted as a time of operation on a full charge. In all, the additional elements describe general equipment, data gathering necessary to perform the calculating step using machine learning, and a step of maintaining a battery range without detailing how the range is maintained.
At Step 2B, when considered as a whole, claim 1 recites in general terms: taking time series data of at least one cell of a battery; calculating an aging state with machine learning that accounts for the interdependencies of multiple parameters of the collected data; and maintaining the battery range using the calculated aging state by adjusting a parameter or factor related to aging state. Therefore claim 1 as a whole does not amount to significantly more than the judicial exception.
Claim 4 is an apparatus claim which measures a battery of a motor vehicle. Claim 4 further recites that the measuring apparatus comprises a sensor, memory unit, and processor, but this is no more limiting than the computing system of claim 1. The apparatus is for a battery of a motor vehicle, which adds a limitation not found in claim 1 and leads one naturally to interpret battery “range” as the distance an electric or hybrid vehicle can travel on a full battery charge. However, because no detail is given on how the battery range is maintained, the same analysis which applies to claim 1 applies to claim 4. Claim 4 is therefore rejected.
Claim 6 is a product claim which specifies that the battery is a motor vehicle battery. This results in the battery “range” being interpreted as described in the rejection of claim 4. Claim 6 also recites a “non-transitory computer readable medium comprising: a computer program code that can be executed on at least one computer, wherein at least one unit of the computer (i) is disposed in a motor vehicle, and/or (ii) is configured to communicate with a cloud in which at least part of the computer program code is provided.” This describes in general terms software and a software product that one would expect to be present for implementing the method described in claims 1 and 6. Furthermore, this limits the computer to either have a component in a motor vehicle, or to be able to communicate with a cloud storing the software. This does not limit the computer to one in a motor vehicle because the computer may in fact be anywhere as long as it can connect to a cloud storing the software. If, however, the software is only stored in the computer, then at least a portion of the computer must be in a motor vehicle. In any case the limitations on the computer does not change the nature of the issues brought up in the rejection of claim 1, therefore claim 6 is similarly rejected.
Claim 2 recites that the calculation in step e. of claim 1 is performed “on a basis of empirical and/or phenomenological data stored in…and/or available to the memory unit.” The scope of “empirical and/or phenomenological data” does not significantly limit how the calculation recited in claim 1 is performed, therefore claim 2 is also rejected.
Claim 5 recites a motor vehicle described in general terms and which comprises the measuring apparatus of claim 4. The issues for which claim 4 was rejected persist when limiting the measuring apparatus of claim 4 to one which is part of an EV, therefore claim 5 is rejected.
Claims 8-10 recite generating a plot of the operating data which depicts parameter interdependencies and known physical measured values. This is insignificant extra-solution activity, therefore claims 8-10 are rejected.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-2, 4-6, and 8-10 are rejected under 35 U.S.C. 103 as being unpatentable over Kusch (US 20120112693 A1) in view of Simonis (US 20220179009 A1; note that foreign application DE 10 2020 215297.7 filed 12/03/2020 discloses all the features of Simonis relied upon in the rejection below) and You (US 20170123009 A1).
Regarding claim 1, Kusch discloses an electric vehicle (EV) containing a traction battery (Abstract; Fig. 1, energy storage system 30; ¶35: energy storage system 30 may include batteries) and an electric drive motor (Figure 1, electric motor 26) but does not teach the method of claim 1.
Simonis discloses a computer-implemented method for determining an aging state of at least one cell of a battery (Abstract: A computer-implemented method for predicting state of health (SOH) of an electrical energy store having at least one electrochemical unit such as a battery cell; ¶37: energy store (i.e. battery) can be used to operate a motor vehicle), said method comprising the following steps:
storing a machine learning architecture in memory of a computing system (¶49: “A data-based state of health model for the respective vehicle battery can be implemented in a control unit in the motor vehicles.” Whatever holds the model could be termed a memory);
training a machine learning architecture for learning a driving behavior of a user of a motor vehicle (¶49: “The state of health model can be continually updated or retrained in a central processing unit based on operating variables of the vehicle batteries from the vehicle fleet.” The operating variables represent driving behavior of users; consider a particular vehicle in the fleet as the “motor vehicle” in the claim language), and for determining the aging state of the at least one cell of the battery of the motor vehicle (the vehicles in the fleet use the state of health model) based on the learned driving behavior (as above, the model can be trained using the driving behavior of all vehicles in the fleet) and historical data related to aging states of at least one cell of other batteries (again, the state of health model is trained on operating variables from other vehicles, and that data is historical because it was measured in the past);
receiving, by at least one sensor of the computing system, operating data relating to the at least one cell of the battery, the operating data including at least: an electrical voltage (¶53: motor vehicle 4 transmits operating variables F of the energy store to a central processing unit (CPU) 2. The variables may include battery current, voltage, temperature, and state of charge (SOC). ¶83 makes clear that at least one sensor collects the operating variables F; the sensor and CPU can be considered parts of a computing system), wherein the operating data is assigned a specified timestamp based on a point of detection (¶53: the data may be time series data captured at a given rate such as 0.1 Hz; as described below with reference to ¶80, a recurrent neural network learns time dependency of the time series data, thus the data is associated with timestamps); and
calculating, by the machine learning architecture (¶18 teaches that the data-based SOH correction model can be a machine learning model, and ¶65 states that the SOH model can be purely data-based and implemented by various machine learning methods), an aging state of the at least one cell of the battery using a processor of the computing system (¶11: a state of health is determined on the basis of the operating variables F; ¶8 the method is implemented on a computer, which is commonly understood to comprise at least a processor and a memory) while interdependently taking into account multiple parameters included in the operating data at common timestamps (¶76: the state of health model may use a usage pattern model 10; ¶80: the usage pattern model 10 uses time series data of the operating variables F and may be a recurrent neural network, thus it accounts for the interdependencies of multiple parameters in the operating data at common timestamps), the aging state calculation using a physical condition of the [at] least one cell of the battery determined from physically measured values (¶53: voltage is collected and used to predict a SOH; a voltage measurement of a battery represents a physical condition of the battery determined from a physically measured value) and the historical data related to the aging states of the at least one cell of other batteries (¶49: the SOH model can be trained on historical operating variables from fleet data, thus it makes use of historical data related to the SOH of other batteries).
While Simonis does not explicitly state that the electrical voltage received is defined across the at least one battery cell, it would have been obvious to define voltage this way as battery voltage is typically defined by the relative electric potential per unit charge between the terminals of a battery or battery cell. Furthermore, while Simonis does not explicitly disclose receiving an electrical charge across the at least one cell of the battery, Simonis does teach receiving SOC (see ¶53), which is the ratio of charge across the at least one cell of a battery relative to the at least one cell’s fully charged state. It would have been obvious, therefore, to receive information on electrical charge itself as an intermediate step to determining SOC.
Simonis does not explicitly disclose that the aging state calculation uses a power output measured at plural common timestamps; however, Simonis does disclose that voltage and current are used (¶53: voltage and current are collected and used to predict a SOH). Since electrical power P=VI, it follows that any two of voltage, current, and power output would provide the same information. Therefore it would have been obvious to one of ordinary skill in the art to use power output in place of voltage or current since one would understand that such an exchange would yield the same information.
Furthermore, Simonis does not explicitly disclose that the aging state calculation uses external conditions determined from the operating data, however Simonis does disclose collecting and using battery temperature (¶53). Simonis also teaches that high ambient temperature can intensify aging (¶23) and that ambient temperature can be derived from weather information or measured directly (¶29). Therefore it would have been obvious to one of ordinary skill in the art to cause the operating data to include information from which external conditions can be derived, then to cause the aging state calculation to use external conditions determined from the operating data. Doing so would be useful since external conditions such as ambient temperature are determinable and are known to affect battery aging.
Finally, Simonis discloses that the technical field is electric vehicle (EV) battery SOH estimation (¶2).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to incorporate the teachings of Simonis with the invention of Kusch by including the method and computing system of Simonis in the EV of Kusch, where the computing system disclosed in Simonis is part of the EV computing system. Doing so would enable one to estimate the SOH of the EV battery, which knowledge can be used to determine if or when the battery requires maintenance or replacement.
Kusch in view of Simonis does not explicitly disclose that the operating data includes torque requests to the electric drive motor. However, Kusch teaches that a torque controller (Fig. 1, torque controller 28) provides control operation of the electric drive motor (¶35: a “torque controller 28 is provided to control operation of electric motor 26”). The torque controller receives commands from an accelerator pedal (Fig. 1, accelerator pedal 38; see ¶35: “Accelerator pedal 38 is configured to send throttle command signals or accelerator pedal signals…torque control 28”). The energy provided to turn the car’s wheels may be provided by the battery via the torque controller (¶38: “energy may be provided to drive shaft assembly 18 via drive control system 28 having energy drawn from energy storage system 30”). Furthermore, Simonis teaches that battery aging depends in part on the load over a battery, which is a function of usage behavior of a driver (¶4: “An extent of the aging of the energy store is dependent on an individual load on the energy store, i.e. in the case of vehicle batteries of motor vehicles on the usage behavior of a driver”).
Noting that a torque request would involve requesting a particular battery load, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to incorporate the teachings of Kusch and Simonis with the invention of Kusch in view of Simonis by causing the received operating data to include torque requests to an electric drive motor. Doing so would enable the SOH calculation model to incorporate data on battery load, which is correlated with battery aging.
Kusch in view of Simonis does not explicitly disclose step d.
You discloses a method for sensing battery data and estimating the state of health of a battery (Abstract). As part of the invention, You discloses a battery manager 134 which may control voltage or current, control a cooling or heating apparatus, prevent over- or undercharging, and perform cell balancing to maintain battery life, based on the state of charge (SOC) and state of health (SOH) (¶45). The SOH may be determined from an SOH estimator 133 or SOH estimation apparatus 200 comprising a model store 220 (¶44; ¶48: SOH estimation apparatus 200 is an example of SOH estimator 133). The model store may be based on a neural network or other machine learning models (¶54). Thus, You teaches utilizing a neural network to determine aging state and adjust one or more parameters or factors relevant to aging state.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to incorporate the teachings of You in view of Kusch and Simonis by maintaining, by the processor of the computing system, a range of the battery using the calculated aging state to model an interdependency of the multiple parameters of the operating data associated with the power output of at least two common timestamps of the plural common timestamps and the physically measured values and adjust one or more of the multiple parameters or factors relevant to the aging state. Doing so would enable one to use machine learning to autonomously make adjustments to prolong battery life, thus maintaining a range of the battery.
While Kusch in view of Simonis and You does not explicitly teach that the timestamp assigned to the operating data is assigned by the at least one sensor, it would have been obvious to cause the sensor to assign the timestamp because the sensor collects the data. Again, while Kusch in view of Simonis and You does not explicitly teach that the operating data is stored in a memory unit of the computing system with the specified timestamp, it would have been obvious to do so as a way to enable the SOH estimator to access the data.
Regarding claim 4, claim 4 recites a measuring apparatus which performs the method outlined in steps a.-f. of claim 1 (occasionally using different syntax, but nevertheless recite the same limitations), and reasons for rejecting these limitations are found in the rejection of claim 1. Claim 4 limits the battery to one of a motor vehicle, which Kusch discloses (see rejection of claim 1). Again, claim 4 recites that the measuring apparatus comprises “at least one sensor for acquiring operating data relating to at least one cell of the battery; at least one memory unit for storing operating data from the at least one sensor device; and at least one processor for processing the operating data.” These elements are recited in claim 1 as part of the computing system, and therefore the rationale for rejecting these limitations are also found in the rejection of claim 1.
Regarding claim 6, Kusch discloses an electric vehicle (EV) containing a traction battery (Abstract; Fig. 1, energy storage system 30; ¶35: energy storage system 30 may include batteries) but does not teach the method of claim 6.
Simonis recites a non-transitory computer readable medium comprising a computer program code that can be executed on at least one computer, wherein by executing the computer program code, the at least one computer is prompted for performing a method of determining an aging state of a battery (Claim 15, noting that the “data processing device” describes a computer). Furthermore, following the reasoning in the rejection of claim 1, if the car of Kusch incorporates the SOH prediction system of Simonis, it would have been obvious for the at least one computer implementing the SOH prediction method to be in the EV (see also Simonis, ¶49: “A data-based state of health model for the respective vehicle battery can be implemented in a control unit in the motor vehicles.”).
Most of the remaining limitations of claim 6 are also found in claim 1 and are rejected for the same reasons. Step a. of claim 6 further limits the at least one sensor to one of a motor vehicle and the battery to one of a motor vehicle; these limitations have been addressed in the rejection of claim 1 (see rejection of claim 1).
Regarding claim 2, Kusch in view of Simonis and You teaches the limitations of claim 1.
As part of the method for predicting battery cell SOH, Simonis further discloses using hybrid SOH models which may incorporate a physical health model based on electrochemical model equations (which represent empirical and/or phenomenological data) and a data-based model (¶17). By doing so a model may obtain a more accurate calculation or prediction of SOH (¶20).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to incorporate the teachings of Simonis with the invention of Kusch in view of Simonis and You by causing the aging state calculation to be carried out on a basis of empirical and/or phenomenological data stored in the memory unit and/or available to the memory unit. By doing so one would be able to incorporate electrochemical model equations as well as collected data for aging state predictions, thus using known empirical and/or phenomenological relationships to obtain more accurate predictions.
Regarding claim 5, Kusch in view of Simonis and You teaches the limitations of claim 4. Kusch further discloses a motor vehicle comprising:
a drive train (Abstract) and at least one drive wheel (Figure 1, wheels 24), wherein the drive train comprises an electric drive motor (Figure 1, electric motor 26) and a traction battery for propulsion of the motor vehicle (Figure 1, energy storage system 30; Paragraph 35 states that energy storage system 30 may include batteries),
wherein the electric drive motor, which is supplied with electrical voltage by the traction battery (Figure 1, energy storage 30, torque controller 28, electric motor 26), is connected to the at least one drive wheel in a torque-transmitting manner (Figure 1, electric motor 26, drive shaft assembly 18, differential 16, wheels 24).
While Kusch does not disclose that the EV comprises the measuring apparatus according to claim 4, it would have been obvious to do so in order to estimate the SOH of the EV battery and determine if or when the EV battery requires maintenance or replacement.
Regarding claims 8-10, Kusch in view of Simonis and You teaches the limitations of claims 1, 4, and 6, respectively.
Simonis further discloses collecting operating data and physically measured values over time and plotting them (Fig. 5c, T represents battery temperature and UT represents ambient temperature; see ¶89). The plot depicts interdependencies between multiple physical variables.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to incorporate the teachings of Simonis with the invention of Kusch in view of Simonis and You by plotting, by the/a processor of the/a computing system, the operating data of multiple timestamps, the plot depicting the interdependency of the multiple parameters of the operating data at each timestamp and physical measured values. Plotting would allow one to visually inspect the operating data and physically measured values to search for interdependencies. Doing so would also provide a visual aid for one attempting to establish a relationship between these data and the aging state.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Duan (US 20150360578 A1) teaches that battery degradation depends on battery chemistry, time, temperature, state-of-charge, and driving behavior (see ¶52).
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 ETHAN WESLEY EDWARDS whose telephone number is (571)272-0266. The examiner can normally be reached Monday - Friday, 7:30am-5pm.
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.
ETHAN WESLEY EDWARDS
Examiner
Art Unit 2857
/E.W.E./ Examiner, Art Unit 2857
/ANDREW SCHECHTER/ Supervisory Patent Examiner, Art Unit 2857