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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 4/16/2026 has been entered.
Response to Amendment
Regarding the rejections under 35 USC §101, amendments made to the claims fail to overcome the rejections. The rejections under 35 USC §101 are maintained as outlined below.
Regarding the rejections under 35 USC §103, amendments made to the claims have necessitated a new grounds of rejection as outlined below.
Response to Arguments
Applicant's arguments filed 6/09/2025 have been fully considered but they are considered moot and not persuasive.
Applicant’s arguments directed toward rejections under 35 USC §103 are considered moot because they are directed toward elements that have not been previously considered and have necessitated a new grounds of rejection.
Regarding rejections under 35 USC §101, Applicant asserts the newly amended elements of claim 8 “cannot be practically performed in the mind… the determining step as recited involves processing signals from physical sensors…” in page 5 of Applicant’s remarks. However, examiner respectfully disagrees. The claim limitations recites “determining comprising identifying whether collision indicators from vehicle sensors confirm an actual accident following the predicted accident” which does not, as Applicant asserts, require processing of signals. The limitation merely requires an accident to be confirmed or identified from vehicle sensors (emphasis added). This would simply require the sensors to gather data and have One having ordinary skill in the art look at the data to determine whether an accident occurred based on the gathered sensor data. For example, the data can be the deployment of an airbag or audio data of a collision and One having ordinary skill in the art would look at the data to determine that a collision did occur.
Applicant further asserts “the present claims address the technical challenge of maintaining accident prediction model accuracy…” and that the claims “do not broadly preempt the idea of retraining machine learning models or even the idea of improving accident prediction. Rather, the claims recite a particular technical solution” in pages 5-7 of Applicant’s Remarks. However, Examiner respectfully disagrees. The abstract idea is not integrated into a practical application and does not recite a particular technical solution, as written. The claims, as written, are still broadly reciting data gathering and transmission of data between a computing device and server. The “updated machine learning model” is not being used and no other action has been recited to further clarify or claim a practical application of the transmitted/received data.
Therefore, Applicant’s arguments regarding rejections under 35 USC §101 are considered not persuasive and the rejection is maintained as outlined below.
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 8-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without adding significantly more.
101 Analysis – Step 1
Regarding Step 1 of the Revised Guidance, it must be considered whether the claims are directed to one of the four statutory classes of invention. In the instant case, claims 1-13 are directed to a method and recites at least one step, claims 14-20 are directed to a system that comprises a processor and a memory. Therefore, claims 1-20 are within at least one of the four statutory categories (process and apparatus).
101 Analysis – Step 2A, Prong 1
Regarding Prong 1 of the Step 2A analysis in the 2019 PEG, the claims are to be analyzed to determine whether they recite a judicial exception.
Independent claim 8 includes limitations that recite an abstract idea (bolded below). Claim 8 recites:
A method comprising:
receiving, at a computing device, event data from a vehicle, wherein the event data is recorded by the vehicle in response to a machine learning model predicting an accident;
determining, by the computing device, whether an accident occurred based on the event data, the determining comprising identifying whether collision indicators from vehicle sensors confirm an actual accident following the predicted accident;
labeling, by the computing device, the event data as associated with an accident or not associated with an accident based on the determining;
transmitting, by the computing device, the labeled event data to a remote server; and
receiving, by the computing device from the remote server, an updated machine learning model trained using the labeled event data.
The examiner submits that the bolded limitations above constitute a “mental process” because under its broadest reasonable interpretation, the claim covers performance of the limitations in the human mind. For example, One of ordinary skill in the art would be able to, with the aid of pen and paper and with the data provided, predict a possible accident and determine whether an accident has occurred and label the data as “accident occurred” or “no accident.” Furthermore, One of ordinary skill in the art would be able to use gathered sensor data to determine whether the data indicates a collision occurred or not.
101 Analysis – Step 2A, Prong 2
Regarding Prong 2 of the Step 2A analysis in the 2019 PEG, the claims are to be analyzed to determine whether the claim, as a whole, integrates the abstract idea into a practical application. As noted in the 2019 PEG, it must be determined whether any additional elements in the claim, beyond the abstract idea, integrate the exception into a practical application in a manner that imposes a meaningful limit on the judicial exception (mental process). The courts have indicated that additional elements merely using a computer to implement an abstract idea, adding insignificant extra solution activity, or generally linking use of a judicial exception to a particular technological environment or field of use do not integrate a judicial exception into a “practical application.”
In the instant application, the additional limitations beyond the above-noted abstract ideas are as follows (where the underlined portions are the “additional limitations” while the bolded portions continue to represent the “abstract idea”):
A method comprising:
receiving, at a computing device, event data from a vehicle, wherein the event data is recorded by the vehicle in response to a machine learning model predicting an accident;
determining, by the computing device, whether an accident occurred based on the event data, the determining comprising identifying whether collision indicators from vehicle sensors confirm an actual accident following the predicted accident;
labeling, by the computing device, the event data as associated with an accident or not associated with an accident based on the determining;
transmitting, by the computing device, the labeled event data to a remote server; and
receiving, by the computing device from the remote server, an updated machine learning model trained using the labeled event data.
The recitation of “computing device,” “a vehicle,” “machine learning model,” and “a remote server” are provided at a high level of generality. Therefore, the additional elements recited fail to provide a specific technology that is integral to the claim. The limitation “event data is recorded by the vehicle” falls under mere data gathering which is considered insignificant extra-solution activity (see MPEP 2106.05(g)). The MPEP states insignificant extra-solution activity fails to integrate a judicial exception (abstract idea) into a practical application. Therefore, the additional elements merely amount to a general application of the abstract idea into a technological environment and mere data gathering. Thus, the claims must be further examined under Step 2B.
101 Analysis – Step 2B
Regarding Step 2B of the Revised Guidance, it must finally be considered whether the claim includes any additional element or combination of elements that provide an inventive concept (i.e., whether the additional element or elements are sufficient to amount to significantly more than the abstract idea).
In the instant application, the additional limitations beyond the above-noted abstract ideas are as follows (where the underlined portions are the “additional limitations” while the bolded portions continue to represent the “abstract idea”):
A method comprising:
receiving, at a computing device, event data from a vehicle, wherein the event data is recorded by the vehicle in response to a machine learning model predicting an accident;
determining, by the computing device, whether an accident occurred based on the event data, the determining comprising identifying whether collision indicators from vehicle sensors confirm an actual accident following the predicted accident;
labeling, by the computing device, the event data as associated with an accident or not associated with an accident based on the determining;
transmitting, by the computing device, the labeled event data to a remote server; and
receiving, by the computing device from the remote server, an updated machine learning model trained using the labeled event data.
The newly underlined additional limitations “receiving… event data,” “transmitting… the labeled event data,” and “receiving… an updated learning model…” are equivalent to transmitting data over a network which has been considered well-understood, routine, and conventional activity by the Courts (see MPEP 2106.05(d)) which falls under insignificant extra-solution activity. Therefore, the additional elements are not sufficient to amount to significantly more than the judicial exception for the same reasons to those discussed above with respect to determining that the claim does not integrate the abstract idea into a practical application. As discussed above, the additional elements amount to nothing more than merely applying the abstract idea into a technological environment, insignificant extra-solution activity, and well-understood, routine, and conventional activity. Hence, the claim is not patent eligible.
Independent claim 14 is parallel in scope to claim 8 and are ineligible for similar reasons.
Specifically regarding claim 14, “processor,” “memory,” “train a machine learning accident prediction model,” and “retrains the machine learning accident prediction model” are also provided at a high level of generality and also merely amounts to the application of the abstract idea into a technological environment. Thus, the elements also fail to integrate the abstract idea into a practical application and does not amount to significantly more than the abstract idea.
Claims 9-10, and 18 are dependent on claims 8, and 14 and inherit the abstract ideas set forth in claims 8, and 14. No other technology or action has been recited in claims 9-10, and 18 to integrate the abstract idea into a practical application nor to amount to significantly more than the abstract idea. Thus, claims 9-10, and 18 also do not confer eligibility on the claimed invention and are ineligible for reasons stated above and for similar reasons to claims8, and 14.
Claim 11, in addition to the abstract idea set forth in claim 8, recites “converting… event data… to a standardized format” which is also an abstract idea that can be performed in the mind. For example, One of ordinary skill in the art would be able to use the data provided, and with the aid of pen and paper, reformat data from a list to a chart. No additional structure or technology has been recited to integrate the abstract idea into a practical application nor amount to significantly more than the abstract idea. Thus, claim 11 also does not confer eligibility on the claimed invention and is ineligible for reasons stated above and for reasons similar to claim 8.
Claim 12, in addition to the abstract idea set forth in claim 8, recites “analyzing the event data to identify false positive predictions and adjusting the machine learning model” which is also an abstract idea that can be performed in the mind. For example, One of ordinary skill in the art would be able to calculate a false positive rate using the data provided and adjust a parameter or weight in a calculation based on the false positive rate with the aid of pen and paper. No additional structure or technology has been recited to integrate the abstract idea into a practical application nor amount to significantly more than the abstract idea. Thus, claim 12 also does not confer eligibility on the claimed invention and is ineligible for reasons stated above and for reasons similar to claim 8.
Claim 13, in addition to the abstract idea set forth in claim 8, recites “adjusting weights” which is also an abstract idea that can be performed in the mind. For example, One of ordinary skill in the art would be able to adjust a parameter or weight in a calculation based on the false positive rate with the aid of pen and paper. No additional structure or technology has been recited to integrate the abstract idea into a practical application nor amount to significantly more than the abstract idea. Thus, claim 13 also does not confer eligibility on the claimed invention and is ineligible for reasons stated above and for reasons similar to claim 8.
Claim 15, in addition to the abstract idea set forth in claim 14, recites “normalize the event data… to a standardized format” which is also an abstract idea that can be performed in the mind. For example, One of ordinary skill in the art would be able to use the data provided, and with the aid of pen and paper, reformat data from a list to a chart. No additional structure or technology has been recited to integrate the abstract idea into a practical application nor amount to significantly more than the abstract idea. Thus, claim 15 also does not confer eligibility on the claimed invention and is ineligible for reasons stated above and for reasons similar to claim 14.
Claim 16, in addition to the abstract idea set forth in claim 8, recites “compute false positive and false negative rates for the event data and adjusts parameters” which is also an abstract idea that can be performed in the mind. For example, One of ordinary skill in the art would be able to calculate a false positive rate and false negative rate using the data provided and adjust a parameter or weight in a calculation based on the false positive rate with the aid of pen and paper. No additional structure or technology has been recited to integrate the abstract idea into a practical application nor amount to significantly more than the abstract idea. Thus, claim 16 also does not confer eligibility on the claimed invention and is ineligible for reasons stated above and for reasons similar to claim 14.
Claim 17 inherits the abstract idea set forth in claim 14 due to dependency. The additional element “re-training the accident model” is provided at a high level of generality and merely amounts to the application of the abstract idea into a technological environment. Therefore, the additional element fails to integrate the abstract idea into a practical application and does not amount to significantly more than the abstract idea. Thus, claim 17 also does not confer eligibility on the claimed invention and is ineligible for reasons stated above and for reasons similar to claim 14.
Claim 19 inherits the abstract idea set forth in claim 14 due to dependency. The additional element “generate a global prediction model…” is general data output and is considered insignificant extra-solution activity. Therefore, the additional element fails to integrate the abstract idea into a practical application and does not amount to significantly more than the abstract idea. Thus, claim 17 also does not confer eligibility on the claimed invention and is ineligible for reasons stated above and for reasons similar to claim 14.
Claim 20 inherits the abstract idea set forth in claim 14 due to dependency. The additional element “store the event data” is equivalent to “storing and retrieving information in memory” which has been considered well-understood, routine, and conventional activity by the Courts (see MPEP 2106.05(d)). Therefore, the additional element fails to integrate the abstract idea into a practical application and does not amount to significantly more than the abstract idea. Thus, claim 20 also does not confer eligibility on the claimed invention and is ineligible for reasons stated above and for reasons similar to claim 14.
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 1-4, 8-10, 14, and 17-18 are rejected under 35 U.S.C. 103 as being unpatentable over Shin et al. (U.S. Patent Application Publication No. 2019/0371087 A1; hereinafter Shin) in view of Sathyanarayana et al. (U.S. Patent Application Publication No. 2018/0293449 A1; hereinafter Sathyanarayana) and Nejah et al. (U.S. Patent No. 10,282,922 B1; hereinafter Nejah).
Regarding claim 1, Shin discloses:
A method comprising: training, by a computing device, an accident model using classified data (neural network model trained to determine whether an event is occurred using driving image obtained during driving, see at least [0104]);
receiving, by the computing device, event data and accident labels from vehicles (server includes an update module 210 which receives event frames from vehicle terminals, see at least [0106]; AI processing unit 180 in vehicle determines whether an event has occurred, the event including an accident, if it is determined that an event has occurred, the AU processing unit 180 extracts the instant as an event frame and is transmitted to the server 20, see at least [0131]-[0132] and [0137]-[0138]);
re-training, by the computing device, the accident model using the event data (after update module 210 of server 200 receives the event frame, the neural network model is trained through reinforcement learning and generates an update file, see at least [0143]-[0144]) ; and
transmitting, by the computing device, the accident model to the vehicles (update file is transmitted to all the vehicle terminals 200 connected to the server and AI processing unit 180 is updated in accordance with the update file received at the vehicle terminal, see at least [0144]-[0145]).
Shin does not explicitly disclose:
accident label
wherein the accident labels comprise post-prediction determinations by the vehicle as to whether actual accidents occurred following predictions of potential accidents by a machine learning model executing on the vehicles
However, Sathyanarayana teaches:
receiving, by the computing device, event data and accident labels from vehicles (at an on-board system mounted to a vehicle, recording video, labeling near-collision event, and transmitting the label to a remote computing system, see at least claim 1)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the collection of learning data and determining whether an event has occurred as disclosed by Shin by adding the transmitting of label event data to a remote system as taught by Sathyanarayana with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order to “reduce or conserve the computational resources and/or power consumed” (see [0025]).
Furthermore, Nejah teaches:
wherein the accident labels comprise post-prediction determinations by the vehicle as to whether actual accidents occurred following predictions of potential accidents by a machine learning model executing on the vehicles (smart crash detector connected to an on board diagnostic system and obtains in-car sensor information, see at least col. 2 lines 6-9; neural network may classify features into categories of crash and non-crash, see at least col. 4 lines 2-3 and claim 1)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the collection of learning data and determining whether an event has occurred as disclosed by Shin and the transmitting of label event data to a remote system as taught by Sathyanarayana by adding the classification of a crash using sensor data taught by Nejah with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order “to improve the effectiveness of emergency services by making accident detection fully automated” (see col. 2 lines 27-29).
Regarding claim 2, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above and Shin further discloses:
the classified data includes black box data recorded by vehicles (vehicle terminal such as black box transmits images of event to server, see at least [0006]).
Regarding claim 3, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above and Shin further discloses:
the event data comprises data recorded by vehicles while operating (event frame is obtained from operation of image obtaining unit 110, see at least [0112]; AI processing unit 180 determines in real time whether event occurs, see at least [0131]).
Regarding claim 4, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above and Shin further discloses:
the accident labels comprises labels predicted by a machine learning model executing on the vehicles (AI processing unit 180 determines event based on installed neural network model to determine traffic accident and violation of traffic regulations, see at least [0112]; neural network may include deep learning model, see at least [0116])
Regarding claim 8, Shin discloses:
A method comprising: receiving, at a computing device, event data from a vehicle, wherein the event data is recorded by the vehicle in response to a machine learning model predicting an accident (server includes an update module 210 which receives event frames from vehicle terminals, see at least [0106]; AI processing unit 180 in vehicle determines whether an event has occurred, the event including an accident, if it is determined that an event has occurred, the AU processing unit 180 extracts the instant as an event frame and is transmitted to the server 20, see at least [0131]-[0132] and [0137]-[0138]);
determining, by the computing device, whether an accident occurred based on the event data (AI processing unit 180 in vehicle determines whether an event has occurred, the event including an accident, see at least [0112]);
transmitting, by the computing device, the labeled event data to a remote server (server includes an update module 210 which receives event frames from vehicle terminals, see at least [0106]; communication network 300 transmits event frame to server, see at least [0142]); and
receiving, by the computing device from the remote server, an updated machine learning model trained using the labeled event data (update file is transmitted to all vehicle terminals connected to server 200 through communication network 300, see at least [0145]).
Shin does not explicitly disclose:
the determining comprising identifying whether collision indicators from vehicle sensors confirm an actual accident following the predicted accident
labeling, by the computing device, the event data as associated with an accident or not associated with an accident based on the determining
However, Sathyanarayana teaches:
labeling, by the computing device, the event data as associated with an accident or not associated with an accident based on the determining (at an on-board system mounted to a vehicle, recording video, labeling near-collision event, and transmitting the label to a remote computing system, see at least claim 1)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the collection of learning data and determining whether an event has occurred as disclosed by Shin by adding the transmitting of label event data to a remote system as taught by Sathyanarayana with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order to “reduce or conserve the computational resources and/or power consumed” (see [0025]).
Furthermore, Nejah teaches:
the determining comprising identifying whether collision indicators from vehicle sensors confirm an actual accident following the predicted accident (smart crash detector connected to an on board diagnostic system and obtains in-car sensor information, see at least col. 2 lines 6-9; neural network may classify features into categories of crash and non-crash, see at least col. 4 lines 2-3 and claim 1)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the collection of learning data and determining whether an event has occurred as disclosed by Shin and the transmitting of label event data to a remote system as taught by Sathyanarayana by adding the classification of a crash using sensor data taught by Nejah with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order “to improve the effectiveness of emergency services by making accident detection fully automated” (see col. 2 lines 27-29).
Regarding claim 9, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above and Shin further discloses:
executing, by the computing device, the updated machine learning model to predict accidents based on sensor data from the vehicle (event frame is obtained from operation of image obtaining unit 110, see at least [0112]; AI processing unit 180 determines in real time whether event occurs, see at least [0131]).
Regarding claim 14, Shin discloses:
A system comprising: a vehicle computing device (vehicle terminal 100, see at least [0099]) comprising a processor and a memory (vehicle terminal 100 includes controller 170 and memory 160, see at least [0109]), the memory storing event data comprising sensor data recorded by the vehicle (memory 160 stores event frame, see at least [0129]); and
a remote server (server 200) configured to: train a machine learning accident prediction model using classified training data (server 200 includes update module 210 uses event frames as learning data to train neural network model, see at least [0106]);
transmit the machine learning accident prediction model to the vehicle computing device (update file is transmitted to all the vehicle terminals 200 connected to the server and AI processing unit 180 is updated in accordance with the update file received at the vehicle terminal, see at least [0144]-[0145]), wherein the vehicle computing device executes the model on the sensor data to predict accidents and records the event data in response to a predicted accident (AI processing unit 180 in vehicle determines whether an event has occurred, the event including an accident, if it is determined that an event has occurred, the AU processing unit 180 extracts the instant as an event frame and is transmitted to the server 20, see at least [0131]-[0132] and [0137]-[0138]); and
receive the event data from the vehicle computing device, determines whether an actual accident occurred, and retrains the machine learning accident prediction model using the event data and accident labels (after update module 210 of server 200 receives the event frame, the neural network model is trained through reinforcement learning and generates an update file, see at least [0143]-[0144]).
Shin does not explicitly disclose:
determine accident labels for the event data indicating whether an actual accident occurred, based on collision indicators detected by vehicle sensors following the predicted accident
However, Sathyanarayana teaches:
determines accident labels for the event data indicating whether an actual accident occurred (at an on-board system mounted to a vehicle, recording video, labeling near-collision event, and transmitting the label to a remote computing system, see at least claim 1)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the collection of learning data and determining whether an event has occurred as disclosed by Shin by adding the transmitting of label event data to a remote system as taught by Sathyanarayana with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order to “reduce or conserve the computational resources and/or power consumed” (see [0025]).
Furthermore, Nejah teaches:
based on collision indicators detected by vehicle sensors following the predicted accident (smart crash detector connected to an on board diagnostic system and obtains in-car sensor information, see at least col. 2 lines 6-9; neural network may classify features into categories of crash and non-crash, see at least col. 4 lines 2-3 and claim 1)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the collection of learning data and determining whether an event has occurred as disclosed by Shin and the transmitting of label event data to a remote system as taught by Sathyanarayana by adding the classification of a crash using sensor data taught by Nejah with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order “to improve the effectiveness of emergency services by making accident detection fully automated” (see col. 2 lines 27-29).
Regarding claim 17, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above and Shin further discloses:
the remote server is configured to periodically (access procedure with vehicle terminals are periodically transmitted, see at least [0015]) retrain the model using aggregated event data and accident labels (after update module 210 of server 200 receives the event frame, the neural network model is trained through reinforcement learning and generates an update file, see at least [0143]-[0144]) from a plurality of vehicles (server connected to a plurality of vehicles, see at least [0014]).
Regarding claim 18, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above and Shin further discloses:
the classified training data comprises black box data extracted from vehicles involved in actual accidents (vehicle terminal such as black box transmits images of event to server, see at least [0006]; black box obtains and stores images when accident occurs, see at least [0003]).
Claims 5-7, 12-13, and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Shin in view of Sathyanarayana and Nejah as applied to claim 1 above and further in view of Singh et al. (U.S. Patent No. 9,690,933 B1; hereinafter Singh).
Regarding claim 5, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above but does not teach:
computing a false positive rate for the event data and accident labels
However, Singh teaches:
computing a false positive rate for the event data and accident labels (detecting consistent false negative events by classification engine may exceed a certain rate, see at least col. 7 lines 29-31)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the neural network model disclosed by Shin, the transmitting of label event data to a remote system as taught by Sathyanarayana, and the classification of a crash using sensor data taught by Nejah by adding the false negative detection taught by Singh with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order “to reduce the number or rate of false negative events by the classification engine” by modifying the predictive model (see col. 2 lines 42-47).
Regarding claim 6, the combination of Shin, Sathyanarayana, Nejah, and Singh teaches the elements above but Shin does not disclose:
re-training the accident model comprises determining that the false positive rate is less than a threshold and re-training the accident model in response
However, Singh teaches:
re-training the accident model comprises determining that the false positive rate is less than a threshold and re-training the accident model in response (in response to detected false negative event exceeding a prescribed rate, an alert is sent to update the classification engine, see at least col. 7 lines 26-34; alert is provided and reference classification 140 updates platform-based classification engines with the reference model, see at least col. 12 lines 12-18)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the neural network model disclosed by Shin, the transmitting of label event data to a remote system as taught by Sathyanarayana, and the classification of a crash using sensor data taught by Nejah by adding the updates based on false negatives taught by Singh with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order “to reduce the number or rate of false negative events by the classification engine” by modifying the predictive model (see col. 2 lines 42-47).
Regarding claim 7, the combination of Shin, Sathyanarayana, Nejah and Singh teaches the elements above but Shin does not disclose:
adjusting at least one parameter or weight of the accident model based on the false positive rate
However, Singh teaches:
adjusting at least one parameter or weight of the accident model based on the false positive rate (once the alert is issued, one or more parameters within a predictive model are modified to reduce the number or rate of false negatives, see at least col. 2 lines 39-47)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the neural network model disclosed by Shin, the transmitting of label event data to a remote system as taught by Sathyanarayana, and the classification of a crash using sensor data taught by Nejah by adding the model parameter modification taught by Singh with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order “to reduce the number or rate of false negative events by the classification engine” by modifying the predictive model (see col. 2 lines 42-47).
Regarding claim 12, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above but does not teach:
labeling the event data comprises analyzing the event data to identify false positive predictions; and adjusting the machine learning model based on the identified false positive predictions
However, Singh teaches:
labeling the event data comprises analyzing the event data to identify false positive predictions; and adjusting the machine learning model based on the identified false positive predictions (decreasing values assigned to certain types of features to reduce the number or rate of false positive events, see at least col. 4 lines 33-35)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the neural network model disclosed by Shin, the transmitting of label event data to a remote system as taught by Sathyanarayana, and the classification of a crash using sensor data taught by Nejah by adding the modification taught by Singh with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification to produce “an updated predictive (reference) model” (see col. 4 line 23).
Regarding claim 13, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above but does not teach:
adjusting the machine learning model comprises one or more of adjusting weights of the model, adjusting an activation function, adjusting a loss function, or changing a model architecture
However, Singh teaches:
adjusting the machine learning model comprises one or more of adjusting weights of the model, adjusting an activation function, adjusting a loss function, or changing a model architecture (once the alert is issued, one or more parameters within a predictive model are modified to reduce the number or rate of false negatives, see at least col. 2 lines 39-47)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the neural network model disclosed by Shin, the transmitting of label event data to a remote system as taught by Sathyanarayana, and the classification of a crash using sensor data taught by Nejah by adding the model parameter modification taught by Singh with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order “to reduce the number or rate of false negative events by the classification engine” by modifying the predictive model (see col. 2 lines 42-47).
Regarding claim 16, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above but does not teach:
the remote server is configured to compute false positive and false negative rates for the event data and adjusts parameters of the machine learning model based on the false positive and false negative rates.
However, Singh teaches:
the remote server is configured to compute false positive and false negative rates for the event data and adjusts parameters of the machine learning model based on the false positive and false negative rates (in response to detected false negative event exceeding a prescribed rate, an alert is sent to update the classification engine, see at least col. 7 lines 26-34; alert is provided and reference classification 140 updates platform-based classification engines with the reference model, see at least col. 12 lines 12-18; decreasing values assigned to certain types of features to reduce the number or rate of false positive events, see at least col. 4 lines 33-35)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the neural network model disclosed by Shin, the transmitting of label event data to a remote system as taught by Sathyanarayana, the classification of a crash using sensor data taught by Nejah by adding the updates based on false negatives taught by Singh with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order “to reduce the number or rate of false negative events by the classification engine” by modifying the predictive model (see col. 2 lines 42-47).
Claims 10 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Shin in view of Sathyanarayana and Nejah as applied to claims 8 and 14 above and further in view of Elkenkamp (U.S. Patent Application Publication No. 2017/0094231 A1).
Regarding claim 10, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above but does not teach:
the event data comprises sensor data from the vehicle recorded over a time window preceding a predicted accident
However, Elkenkamp teaches:
the event data comprises sensor data from the vehicle recorded over a time window preceding a predicted accident (video file captured and stored can encompass period of time beginning prior to the sensed event, see at least [0014])
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the event extraction disclosed by Shin, the transmitting of label event data to a remote system as taught by Sathyanarayana, and the classification of a crash using sensor data taught by Nejah by adding the buffer storage taught by Elkenkamp with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order “to capture and store only a portion of the video file related to a specific event” (see [0014]).
Regarding claim 20, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above but does not teach:
the vehicle computing device comprises an event data buffer that stores the event data comprising sensor data recorded over a time window preceding the predicted accident.
However, Elkenkamp teaches:
the vehicle computing device comprises an event data buffer (portion of video stream can be temporarily retained in the buffer storage 115, see at least [0015]) that stores the event data comprising sensor data recorded over a time window preceding the predicted accident (video file captured and stored can encompass period of time beginning prior to the sensed event, see at least [0014])
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the event extraction disclosed by Shin, the transmitting of label event data to a remote system as taught by Sathyanarayana, and the classification of a crash using sensor data taught by Nejah by adding the buffer storage taught by Elkenkamp with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order “to capture and store only a portion of the video file related to a specific event” (see [0014]).
Claims 11 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Shin in view of Sathyanarayana and Nejah as applied to claims 8 and 14 above and further in view of Madau et al. (U.S. Patent Application Publication No. 2011/0055292 A1; hereinafter Madau).
Regarding claim 11, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above but does not teach:
converting, by the computing device, the event data from a vehicle-specific format to a standardized format prior to transmitting the labeled event data
However, Madau teaches:
converting, by the computing device, the event data from a vehicle-specific format to a standardized format prior to transmitting the labeled event data (converting extracted data to a standardized format, see at least [0018])
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the server communications disclosed by Shin, the transmitting of label event data to a remote system as taught by Sathyanarayana, and the classification of a crash using sensor data taught by Nejah by adding the conversion into standardized format taught by Madau with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification “in order to provide compatibility between the gateway module 12 and the protocol used by the vehicle network 14” (see [0015]).
Regarding claim 15, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above but does not teach:
the remote server is configured to normalize the event data from a vehicle-specific format to a standardized format prior to retraining the model
However, Madau teaches:
the remote server is configured to normalize the event data from a vehicle-specific format to a standardized format prior to retraining the model (converting extracted data to a standardized format, see at least [0018])
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the server communications disclosed by Shin, the transmitting of label event data to a remote system as taught by Sathyanarayana, and the classification of a crash using sensor data taught by Nejah by adding the conversion into standardized format taught by Madau with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification “in order to provide compatibility between the gateway module 12 and the protocol used by the vehicle network 14” (see [0015]).
Claim 19 is rejected under 35 U.S.C. 103 as being unpatentable over Shin in view of Sathyanarayana and Nejah as applied to claim 14 above and further in view of Slavin (U.S. Patent No. 10,997,430 B1).
Regarding claim 19, the combination of Shin, Sathyanarayana, and Nejah teaches the elements above but does not teach:
the remote server is configured to generate a global prediction model and separate sub-models customized for different vehicle makes and models using the event data and labels from the different vehicle makes and models.
However, Slavin teaches:
the remote server is configured to generate a global prediction model and separate sub-models customized for different vehicle makes and models using the event data and labels from the different vehicle makes and models (pattern recognition module 206 uses stored event data to label data which is used to train vehicle driver models to generate models for particular vehicle make/model, see at least col. 13 line 61 – col. 14 line 2)
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the AI model disclosed by Shin, the transmitting of label event data to a remote system as taught by Sathyanarayana, and the classification of a crash using sensor data taught by Nejah by adding the particular model taught by Slavin with a reasonable expectation of success. One of ordinary skill in the art would have been motivated to make this modification in order “to detect and identify the repeat offenders who may be frequently reported as driving dangerously” (see col. 14 lines 9-11).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Siddiqui (U.S. Patent Application Publication No. 2019/0111876 A1) teaches a smart surface for detecting collision forces using a neural network to determine that an accident has occurred using the sensor data.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HANA LEE whose telephone number is (571)272-5277. The examiner can normally be reached Monday-Friday: 7:30AM-4:30PM EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jelani Smith can be reached at (571) 270-3969. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/H.L./Examiner, Art Unit 3662
/DALE W HILGENDORF/Primary Examiner, Art Unit 3662
Prosecution Insights