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 .
Status of Claims
Applicant’s submission filed on 04/07/2026 has been entered. The status of claims is as follows:
Claim 1-20 remains pending in the application.
Claims 7 and 16 are cancelled.
Response to Arguments
In reference to the Claim Rejections under 35 U.S.C 103:
Argument 1:
Applicant argues in Remarks pg. 9-12 that the Office Action does not identify any reference that teaches a server communicating a likelihood of trip start score to a mobile device for use in scaling a pre-trip awake window, and that the Office Action also improperly relies on Lehmann by using Examiner’s own bracketed interpretation rather than an express teaching from the reference. Specifically, Applicant contends that Lehmann’s “commuter mode”, “instructed driving mode”, and “silent mode” merely describe post-vehicle-start navigation/ display modes for an already operating navigation device, not a mobile device receiving a likelihood score from a remote server, staying awake for a variable window based on that score, or transitioning back to sleep if no trip data is received. Applicant further argues that the use of “e.g.,” in the Office Action shows that the cited disclosure is being analogized or recharacterized, rather than clearly and unequivocally disclosing the claimed limitations.
Response 1:
Applicant’s argument has been fully considered, but is not persuasive. Examiner respectfully acknowledges that the limitation was inadvertently omitted from the discussion in the Office Action mailed on 12/09/2025 due to a typographical error. However, the record shows that this limitation was previously addressed and rejected in the Office Action mailed on 01/16/2025. In particular, the 01/16/2025 Office Action mapped the limitation to Spears. Specifically, Spears discloses that the server predicts the time period and route of the next travel of the mobile device, uses that prediction to select location-based information, and schedules the information for delivery to the mobile device a predetermined time period before the mobile device starts to travel along the predicted route. Spears further discloses that the predicted route and trip time may be used with utility scores and route history including route frequencies and traveled time windows. Thus, Spears teaches communicating to the mobile device information indicating the predicted trip start time and a score-based indication associated with the likelihood that the trip will occur by that predicted time. Examiner has made a correction to show the mappings of Spears below.
Applicant’s argument that the Examiner relied on an improper personal interpretation is not persuasive. The bracketed discussion in the Office Action was not intended to replace or modify the disclosure of Lehmann, but rather to explain the Examiner’s interpretation of how the cited paragraphs correspond to the claimed limitations under the broadest reasonable interpretation. The use of “e.g.,” merely clarifies the Examiner’s mapping and does not constitute an admission that the rejection is based on unsupported characterization rather than the cited disclosure.
Argument 2:
Applicant argues in Remarks pg. 12-13 that the Office Action improperly relies on the Examiner’s recharacterization of Lehmann rather than Lehmann’s actual disclosure. Specifically, Applicant contends that Lehmann merely teaches prediction-score thresholds used to select different post-vehicle-start navigation modes, such as commuter mode, instructed driving mode, or silent mode, and does not teach a likelihood-score-scaled awake window or automatic sleep transition before a trip begins. Applicant further argues that equating Lehmann’s high-confidence commuter mode with a mobile device remaining wake for a longer period improperly supplies a missing pre-trip sleep/ wake power-management concept that is not disclosed in Lehmann or the cited references, and therefore fails to establish a prima facie case of obviousness.
Response 2:
Applicant’s argument has been considered, but is not persuasive to the extent it asserts that the Office Action relied solely on an unsupported recharacterization of Lehmann. Lehmann expressly discloses that the navigation device supports multiple operational modes, including an instructed driving mode, commuter mode, and silent mode, and that these modes are selected based on prediction score thresholds. In particular, Lehmann discloses that when the prediction score is below a second threshold, the navigation device switches to “silent mode” and does not provide the driver with instructions or present predicted destinations. Lehmann further discloses that the device remains in silent mode until at least one destination can be predicted with a prediction score higher than the applicable threshold, at which point the device switches to another operational mode. Thus, the cited paragraphs of Lehmann are relied upon their express teaching of transitioning the device between different operational states based on whether sufficient trip prediction information is available. The Examiner’s bracketed explanation was provided only to clarify how the cited disclosure corresponds to the claimed limitation under the BRI, not to substitute a new teaching for Lehmann’s disclosure.
Argument 3:
Applicant argues in Remarks pg. 13-14 that the Office Action fails to identify any reference that teaches communicating an indication of the predicted trip start time to the mobile device.
Response 3:
Applicant’s argument has been fully considered, but is not persuasive. Examiner respectfully acknowledges that the limitation was inadvertently omitted from the discussion in the Office Action mailed on 12/09/2025 due to a typographical error. However, the record shows that this limitation was previously addressed and rejected in the Office Action mailed on 01/16/2025. In particular, the 01/16/2025 Office Action mapped the limitation to Spears. Specifically, Spears discloses that the server predicts the time period and route of the next travel of the mobile device, uses that prediction to select location-based information, and schedules the information for delivery to the mobile device a predetermined time period before the mobile device starts to travel along the predicted route. Spears further discloses that the predicted route and trip time may be used with utility scores and route history including route frequencies and traveled time windows. Thus, Spears teaches communicating to the mobile device information indicating the predicted trip start time and a score-based indication associated with the likelihood that the trip will occur by that predicted time. Examiner has made a correction to show the mappings of Spears below.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The 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, 9-11, 13, 15, and 18-19 are rejected under 35 U.S.C. 103 as being unpatentable over Yoshikawa et al. (US PGPUB No US2005/0143905 A1) (hereafter referred to as Yoshikawa) in view of Lehmann et al. (US PGPUB No US2011/0238289 A1) (hereafter referred to as Lehmann), Makoto (US PGPUB No US2005/0288850 A1) (hereafter referred to as Makoto), Spears (US 2015/0264532 A1) and in further view of Loriaux (US 2020/0126123 A1).
Regarding claim 1, Yoshikawa discloses a computing platform:
comprising: at least one processor; a communication interface communicatively coupled to the at least one processor, and memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to: receive historical trip data corresponding to driving trip patterns (Yoshikawa: “FIG. 1 is a conceptual diagram of a navigation apparatus according to an exemplary embodiment of the invention. FIG. 1 shows an information providing server 11 implemented, for example, in a computer that may include a controller such as, for example, a CPU or an MPU, storage means such as a semiconductor memory, a magnetic disk, or an optical disk, and a communication interface.” [0021][Claim 19: “A storage medium storing a set of program instructions executable on a data processing device”] … “The life pattern memory 14 (e.g., a database) stores life pattern data produced by extracting life patterns from daily driving data and/or life information associated with the user, such as schedule information, received from the navigation apparatus 31 or the information terminal 32 of the user.” [*Examiner note: emphasis added. The life patterns from daily driving data are being interpreted as historical trip data] [0034]);
receive initial data corresponding to a particular individual; input the initial data corresponding to a particular individual into the machine learning model (Yoshikawa: “Various exemplary embodiments provide a navigation system including a controller that collects life information associated with a user” [0009] … “The life information processing unit 26 can extract a life pattern from the life information and can register the extracted life pattern. The extracted life pattern may be stored, for example, in the life pattern memory 14. The prediction unit 27 can determine, for example, a predicted start time and a predicted destination on the basis of the life pattern information” [*Examiner note: i.e., the prediction unit 27 is being interpreted as a machine learning model] [0038] …. “First, the information providing server 11 waits for life information to be uploaded from the navigation apparatus 31 or the information terminal 32 (Step S31). … the information providing server 11 extracts a life pattern, by example, performing statistical processing on the received life information (Step S32)” [0058]),
wherein inputting the initial data corresponding to the particular individual into the machine learning model causes output of a predicted start time of a driving trip of the particular individual based at least in part the prior location data (Yoshikawa: “The life information processing unit 26 can extract a life pattern from the life information and can register the extracted life pattern. The extracted life pattern may be stored, for example, in the life pattern memory 14. The prediction unit 27 can determine, for example, a predicted start time and a predicted destination on the basis of the life pattern information” [*Examiner note: emphasis added i.e., the prediction unit 27 is being interpreted as a machine learning model] [0038]);
send, to a mobile device corresponding to the particular individual one or more commands directing the mobile device to wake up prior to the predicted start time and to initiate collection of driving trip data corresponding to the driving trip, wherein directing the mobile device to wake up causes the mobile device to be configured for the collection of driving trip data (Yoshikawa: “FIG. 1 shows a navigation apparatus 31 that may be used by a user (i.e., a mobile device) … The user may be, for example, a driver or a passenger of a vehicle such as a car, a truck, a bus, or a motorcycle” [0022] … “The navigation apparatus 31 can include a controller, such as, for example, a CPU or an MPU, a storage device … a display … an input device such as a keyboard … or a touch panel” [0023] … “If it is determined that it is X minutes before the predicted start time (step S42=YES), the information providing server 11 determines whether the vehicle of the user is parked within an area in which communication with the access point is possible” [0065] … “if it is determined that the vehicle of the user is parked within an area in which communication with the access point is possible (step S42=YES), the information providing server 11 transmits a start command or signal to the navigation apparatus 31 … In response to the start command, the navigation apparatus is turned on, and thus it becomes possible for the navigation apparatus to receive data. The information providing server 11 transmits the predicted congestion information to the navigation apparatus 31 via the access point (step S45)” [*Examiner note: emphasis added] [0066]).
receive the driving trip data from the mobile device; input the driving trip data into the machine learning model (Yoshikawa: “Now the process performed by the information providing server 11 to extract a life pattern from the uploaded life information is described with reference to FIG. 5. FIG. 5 is a flow chart showing a process of extracting a life pattern according to an embodiment of the present invention. First, the information providing server 11 waits for life information to be uploaded from the navigation apparatus 31 or the information terminal 32 (step S31). If the information providing server 11 determines that life information has been received (step S31=YES), the information providing server 11 extracts a life pattern by, for example, performing statistical processing on the received life information (step S32)” [0058][FIG. 5]),
wherein inputting the driving trip data into the machine learning model causes output of a predicted trip end time of a driving trip of the particular individual (Yoshikawa: “The life information processing unit 26 can extract a life pattern from the life information and can register the extracted life pattern. The extracted life pattern may be stored, for example, in the life pattern memory 14. The prediction unit 27 can determine, for example, a predicted start time and a predicted destination on the basis of the life pattern information” [*Examiner note: i.e., the prediction unit 27 is being interpreted as a machine learning model] [0038] … “Additionally, a predicted arrival time may be used instead of the predicted start time, and a start point may be used instead of the destination. According to various exemplary embodiments, both the predicted start time and the predicted arrival time may be used” [0064]);
to conserve power (Yoshikawa, ¶[0024]: “In this case, in order to minimize the power consumption, it may be desirable to activate only a part of the navigation apparatus 31 necessary for receiving data.”)
Yoshikawa fails to teach:
training a machine learning model using the historical data corresponding to the driving trip patterns,
wherein the initial data includes prior location data indicating a location of the particular individual prior to entering a vehicle
communicate, to a mobile device corresponding to the particular individual and prior to the particular individual entering the vehicle, an indication of the predicted trip start time corresponding to a time proximate to the particular individual entering the vehicle, a likelihood of trip start score indicative of a likelihood that a trip will start by the predicted trip start time, and
one or more commands that cause the mobile device to wake up prior to the predicted trip start time and to remain awake for a predetermined amount of time associated with the likelihood of trip start score to facilitate collection, by the mobile device, of driving trip data corresponding to the driving trip,
wherein when driving trip data is not received by the mobile device within the predetermined amount of time, the mobile device transitions to a sleep state to conserve power, and
wherein a length of the predetermined amount of time increases with an increase in the likelihood that the trip will start by the predicted trip start time
communicate, to the mobile device, one or more commands directing the mobile device to stop collection of the driving trip data at the predicted trip end time, wherein sending the one or more commands directing the mobile device to stop collection of the driving trip data causes the mobile device to stop collection of the driving trip data at the predicted trip end time.
However, Lehmann teaches:
train a machine learning model using the historical data corresponding to the driving trip patterns (Lehmann: “the data storage comprises a trip history, which is a data structure comprising a multiple trip data objects. Each trip data object is a data object representing a trip that has been executed in the past” [0011] … “The trip history is used by a software component in the following referred to as a ‘learning module’ … the learning model, by training or re-training an existing machine learning algorithm on the trip history” [0014] … “According to a preferred embodiment of the invention, the default operation mode of the navigation device is to predict the destination after every start of the vehicle and to learn by adding new trip data objects to the trip history after each trip and to retrain the machine learning algorithm on a regular basis” [0028] … “In step 503, a mean squared error signal being indicative of the accuracy of the destination prediction is calculated. The weights of the starting parameters are adapted in each layer of the network by the backpropagation algorithm to minimize a mean squared error value. The mean squared error signal value is propagated backward through the network, thereby changing the weights in each layer in a way minimizing the error value … The re-trained and improved version of the destination prediction algorithm 207 is returned in step 504 from the learning module 206 to the destination prediction module 208” [0103][FIG. 5]);
wherein when driving trip data is not received by the mobile device within the predetermined amount of time, the mobile device transitions to a sleep state to conserve power, and (Lehmann, ¶[0044]: “According to preferred embodiments of the invention, at least three operation modes of the navigation device are supported: an 'instructed driving mode', a 'commuter mode and a 'silent mode' and at least a first and a second threshold variable for the prediction score are specified by a user or manufacturer of the navigation device.”, ¶[0051]: “According to some embodiments of the invention, a second prediction score threshold variable is in the range between 0 and 50%. In case the prediction score of all predicted destinations is below said second threshold, the navigation device switches to 'silent mode' and does not provide the driver with instructions on where to drive to or present him predicted destinations.”, ¶[0052]: “In 'silent mode', the navigation device remains silent until, at some point along the route actually chosen by the driver, at least one destination can be predicted with a prediction score higher than the first or the second prediction score threshold variable. In this case, the navigation device will switch to another operation mode.”) [Examiner’s note: “driving trip data is not received by the mobile device within the predetermined amount of time” e.g., “the navigation device does not provide the driver with instructions on where to drive or present him predicted destinations”]
wherein a length of the predetermined amount of time increases with an increase in the likelihood that the trip will start by the predicted trip start time (Lehmann, ¶[0045]: “The 'commuter mode' is a mode of operation wherein the navigation device has predicted at least one destination with high prediction score (high probability value for a particular destination given a high or at least a sufficient prediction algorithm accuracy) and wherein said prediction score lies above a first threshold variable. According to some embodiments of the invention, said first prediction score threshold variable is in the range of 70- 100%. The 'commuter mode' is a mode of operation of the navigation device according to which all those functions of the navigation device are enabled which provide a beneficial effect for a user who is familiar with a particular route.”, ¶[0047]: “In case the calculated prediction score for all predicted destinations is lower than the first threshold variable, but at least one predicted destination has a prediction score being higher than a second prediction score threshold variable, the navigation device is operated in an 'instructed driving mode'. In this mode, functions which require a high reliability of the predictions are disabled or at least not executed automatically.”, ¶[0051]: “According to some embodiments of the invention, a second prediction score threshold variable is in the range between 0 and 50%. In case the prediction score of all predicted destinations is below said second threshold, the navigation device switches to 'silent mode' and does not provide the driver with instructions on where to drive to or present him predicted destinations.”) [Examiner’s note: The highlight indicates that if the device determines that the likelihood of trip prediction score exceeds a first predetermined threshold (e.g., 70 – 100%), the mobile device remains in a “commuter mode” (e.g., remains awake) with all the functions of the navigation device enabled and if the mobile device determines that the likelihood of prediction score does not exceed the first predetermined threshold (e.g., a second prediction score threshold is in the range between 0-50%), the navigation device switches to a “instructed driving mode” with some functions are disabled (e.g., the device remains awake for a shorter period of time than the first period of time)]
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa and Lehmann. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data. Lehmann discloses training a machine learning model on driver data. One of ordinary skill would have motivation to combine Yoshikawa and Lehmann to train the statistical processing model taught by Yoshikawa with the machine learning training taught by Lehmann because utilizing a machine learning model trained on trip history improves the accuracy of the destination prediction. (Lehmann: “The learning module, by training or retraining an existing machine learning algorithm on the trip history, generates a destination prediction algorithm whose prediction accuracy improves over time” [0015])
However, Loriaux explicitly discloses:
receive initial data corresponding to a particular individual, wherein the initial data includes prior location data indicating a location of the particular individual prior to entering a vehicle (Loriaux, ¶[0008]: “When executed by the one or more processors, the instructions cause the computing system to receive user data indicative of (i) a plurality of locations of a user while the user was driving and (ii) times at which the user was driving at the plurality of locations, and store the received user data in the database. The instructions also cause the computing system to provide, at a notification time that is at or prior to the time when the user is likely to begin driving the route, a notification to a mobile device of the user, the notification indicating one or both of (i) the one or more entities and (ii) locations of the one or more entities.”)
communicate, to a mobile device corresponding to the particular individual and prior to the particular individual entering the vehicle, an indication of the predicted trip start time corresponding to a time proximate to the particular individual entering the vehicle, (Loriaux, ¶[0008]: “The instructions also cause the computing system to determine, by analyzing the user data stored in the database, a route that the user is likely to drive and a time when the user is likely to begin driving the route. The instructions also cause the computing system to identify one or more entities each having a location proximate to the route, at least in part by analyzing, for each of the one or more entities, (i) a current price of a good or service provided by the entity and (ii) current prices of goods or services provided by one or more other entities. The instructions also cause the computing system to provide, at a notification time that is at or prior to the time when the user is likely to begin driving the route, a notification to a mobile device of the user, the notification indicating one or both of (i) the one or more entities and (ii) locations of the one or more entities.”)
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa and Loriaux. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. One of ordinary skill would have motivation to combine Yoshikawa and Loriaux to enable the individual to coordinate their arrival at the vehicle more efficiently, reducing idle time and unnecessary waiting. This early notification can prompt the individual to complete pre-departure tasks, prepare personal items, or navigate toward the vehicle in a timely manner. Additionally, aligning the individual’s arrival with the predicted trip start helps optimize vehicle readiness functions, thereby improving energy usage and overall user experience.
However, Spears explicitly discloses:
communicate, to a mobile device corresponding to the particular individual, an
indication of the predicted trip start time, a likelihood of trip start score indicative of a
likelihood that a trip will start by the predicted trip start time, and (Spears, ,i [0033]:
"In FIG. 1, the server (501) predicts the time period and the route of the next travel Qf.
the mobile device ( 405) based on the route that is recently traversed by the mobile
device (405), uses the prediction to select location-based information (503), and
schedules the location-based information (503) for delivery to the mobile device ( 405)
a predetermined time period before the mobile device (405) starts to travel along the
predicted route in accordance with the predicted time period of travel. Thus, the user of
the mobile device ( 405) is provided with an optimal time window to review the
information and arrange the travel using the information.", ,i[0042]: "The predicted
route and time of the trip can be used to select and/or present offers along or near the
route, based on utility scores of the offers that are a function of the closet to the route
as well as most likely to be redeemed.", ,i[0052]: "In FIG. 2, the route dictionary (232)
includes a list of routes ( 401, ... , 403) of past trips of a user (101 ), the frequencies (411,
... , 413) of the user (101) traveling along the routes (401, ... , 403), and the times traveled (421, ... , 423) (e.g., time windows within a day, week, and/or month).")
one or more commands that cause the mobile device to wake up prior to the
predicted trip start time and to remain awake for a predetermined amount of time associated with the likelihood of trip start score to facilitate collection, by the mobile device, of driving trip data corresponding to the driving trip, (Spears, ¶[0033]: “the server (501) predicts the time period and the route of the next travel of the mobile device ( 405) based on the route that is recently traversed by the mobile device (405), uses the prediction to select location-based information (503), and schedules the location-based information (503) for delivery to the mobile device ( 405) a predetermined time period before the mobile device (405) starts to travel along the predicted route in accordance with the predicted time period of travel. Thus, the user of the mobile device ( 405) is provided with an optimal time window to review the information and arrange the travel using the information.”, ¶[0040]: “For example, route candidates can be selected from the route dictionary (232) based on whether the time periods of the route are within a predetermined time window (e.g., next hour)”, ¶[0045]: “One or more top ranked offer candidates can be selected for presentation to the user at a predetermined time prior to the next trip.”) [Examiner’s note: The highlight indicates that the mobile device stays awake for a predetermined amount of time associated with the trip start time. The fact that one or more top ranked route selection candidates being presented to the user via the mobile device at a predetermined time prior to the next trip indicates that the mobile device remains awake during that time]
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa and Spears. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. One of ordinary skill would have motivation to combine Yoshikawa and Spears to ensure the mobile device responsible for data collection is active at the right time while optimizing power consumption. By associating the likelihood of a trip start with a dynamically adjustable “wake-up” period, this approach minimizes unnecessary device activity, extending battery life and reducing energy waste.
However, Makoto discloses:
communicate, to the mobile device, one or more commands directing the mobile device to stop collection of the driving trip data at the predicted trip end time, wherein sending the one or more commands directing the mobile device to stop collection of the driving trip data causes the mobile device to stop collection of the driving trip data at the predicted trip end time (Makoto: “In the interval driving data storing unit 101, driving data of a vehicle, which is a collection of sampling data sampled at predetermined period (for example, 200 msec) in certain driving interval, is stored. The driving interval has certain time range (for example, 1 minute or 10 seconds etc.) from start point time to end point time.” [*Examiner note: emphasis added. i.e., the start time and end time of the interval would be the predicted start time and the predicted end time of the Yoshikawa reference, and collecting data during an interval would result in data collection stopping at the end of the interval (i.e., at the predicted end time)] [0029]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Lehmann and Makoto. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data, Lehmann discloses training a machine learning model on driver data. Makoto discloses collecting data over an interval from a start time to an end time. One of ordinary skill would have motivation to combine Yoshikawa in view of Lehmann and Makoto to replace the end time of the interval for data collection taught by Makoto with the predicted trip end time disclosed by Yoshikawa in view of Lehmann because useful driver information can be determined by only gathering data during an interval of time (Makoto: “It is therefore important to evaluate a driving result in a time interval from a past relative short time to the present time point by using only interval driving data as driving data in the interval and immediately provide the evaluation result to the driver.” [0006]).
Regarding claim 2, Yoshikawa in view of Lehmann and in further view of Makoto, Spears, Loriaux discloses all of the limitations of claim 1 as shown in the rejection above. Lehmann also discloses:
wherein receiving the historical data comprises receiving one or more of: global positioning system (GPS) data, time information, demographics information, income information, accelerometer data, gyroscope data, barometer data, magnetometer data, or social media data (Lehmann: Starting parameters are gathered in steps 301-304 as described previously … A type of starting parameter could be, for example, “current outdoor temperature”, “time”, “day in week” (i.e., time information) [0102]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Makoto, Spears and Lehmann. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data, and Makoto discloses collecting data over an interval from a start time to an end time. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. Lehmann discloses training a machine learning model on driver data. One of ordinary skill would have motivation to combine Yoshikawa in view of Makoto, Spears and Lehmann to include time information in the historical data as taught by Lehmann in the training data disclosed by Yoshikawa in view of Makoto and Spears because this additional time information improves the accuracy of the machine learning model’s predictions (Lehmann: “The accuracy of the destination prediction algorithm can be further refined by taking into consideration in addition the date and date related information. For example, the destination chosen by a driver may strongly depend on the question, if the day is a working day, Saturday or Sunday or Holiday” [0032]).
Regarding claim 4, Yoshikawa in view of Lehmann and in further view of Makoto, Spears, Loriaux discloses all of the limitations of claim 1 as shown in the rejection above. Lehmann also discloses:
wherein the historical data is labelled based on a corresponding historical driving trip (Lehmann: The trip history is a data structure, e.g. a data object of an object-oriented programming language or a table in a relational data base or any other application structure operable to store a set of trip data objects 201 representing past trips. Each trip data object comprises at least a destination information 205 and a set of (i.e., are labeled by) starting parameters 204, wherein the set of starting parameters comprise at least the actual destination of a trip, its starting location, and its starting time and date” [0074]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Makoto, Spears, Loriaux and Lehmann. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data, and Makoto discloses collecting data over an interval from a start time to an end time. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. Lehmann discloses training a machine learning model on driver data. One of ordinary skill would have motivation to combine Yoshikawa in view of Makoto, Spears, Loriaux and Lehmann to include time information in the historical data as taught by Lehmann in the training data disclosed by Yoshikawa in view of Makoto and Spears because this additional time information improves the accuracy of the machine learning model’s predictions (Lehmann: “The accuracy of the destination prediction algorithm can be further refined by taking into consideration in addition the date and date related information. For example, the destination chosen by a driver may strongly depend on the question, if the day is a working day, Saturday or Sunday or Holiday” [0032]).
Regarding claim 6, Yoshikawa in view of Lehmann, Spears, Loriaux and in further view of Makoto discloses all of the limitations of claim 1 as shown in the rejection above. Yoshikawa also discloses:
wherein the mobile device is not configured to collect driving data prior to waking up (Yoshikawa: “communication between a navigation apparatus and a server only starts after the navigation apparatus is turned on. Therefore, data is not received from the server and is not displayed on a display means to a user until the navigation apparatus is turned on” [0007]).
Regarding claim 8, Yoshikawa in view of Lehmann, Spears, Loriaux and in further view of Makoto discloses all of the limitations of claim 1 as shown in the rejection above. Lehmann also discloses:
wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to: dynamically update the machine learning model based on the driving trip data (Lehmann: “The weights of the starting parameters are adapted in each layer of the network by the back-propagation algorithm to minimize a mean squared error value. The mean squared error signal value is propagated backward through the network, thereby changing the weights in each layer in a way minimizing the error value. Provided the driver shows a steady driving behavior regarding his chosen destination, the prediction accuracy will increase upon every added new trip data object and every re-training of the neural network on the increased set of trip data objects” [0060][Figure. 5 step 503]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Makoto, Spears, Loriaux and Lehmann. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data, and Makoto discloses collecting data over an interval from a start time to an end time. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. Lehmann discloses training a machine learning model on driver data. One of ordinary skill would have motivation to combine Yoshikawa in view of Makoto, Spears, Loriaux and Lehmann to continuously update the model as taught by Lehmann using the training data disclosed by Yoshikawa in view of Makoto and Spears because updating the machine learning model trained on trip history improves the accuracy of the destination prediction. (Lehmann: “The learning module, by training or retraining an existing machine learning algorithm on the trip history, generates a destination prediction algorithm whose prediction accuracy improves over time” [0015]).
Regarding claim 9, Yoshikawa in view of Lehmann, Spears, Loriaux and in further view of Makoto discloses all of the limitations of claim 1 as shown in the rejection above. Lehmann also discloses:
wherein the historical data further corresponds to one or more of: car travel patterns, bus travel patterns, boat travel patterns, train travel patterns, bike travel patterns, or motorcycle travel patterns (Lehmann: “the data storage comprises a trip history, which is a data structure comprising a multiple trip data objects. Each trip data object is a data object representing a trip that has been executed in the past [0011] … the driving behaviour revealed by the trip history” [0053] … “The term ‘instructed driving’ or ‘commuter mode’ does not imply that the vehicle has to be a car. Other embodiments of the invention are adapted to the requirements e.g. of bicyclers and pedestrians” [0094]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Makoto, Spears, Loriaux and Lehmann. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data, and Makoto discloses collecting data over an interval from a start time to an end time. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. Lehmann discloses training a machine learning model on driver data. One of ordinary skill would have motivation to combine Yoshikawa in view of Makoto, Spears, Loriaux and Lehmann to train the statistical processing model taught by Yoshikawa in view of Makoto and Spears with the machine learning training taught by Lehmann because utilizing a machine learning model trained on trip history improves the accuracy of the destination prediction. (Lehmann: “The learning module, by training or retraining an existing machine learning algorithm on the trip history, generates a destination prediction algorithm whose prediction accuracy improves over time” [0015]).
Regarding claim 10, Yoshikawa discloses a method:
comprising: at a computing platform comprising at least one processor, a communication interface and memory: (Yoshikawa: “FIG. 1 is a conceptual diagram of a navigation apparatus according to an exemplary embodiment of the invention. FIG. 1 shows an information providing server 11 implemented, for example, in a computer that may include a controller such as, for example, a CPU or an MPU storage means such as a semiconductor memory, a magnetic disk, or an optical disk and a communication interface” [0021][FIG. 1 shows communicative coupling]);
receiving historical data corresponding to driving trip patterns (Yoshikawa: “The life pattern memory 14 (e.g., a database) stores life pattern data produced by extracting life patterns from daily driving data and/or life information associated with the user, such as schedule information, received from the navigation apparatus 31 or the information terminal 32 of the user.” [*Examiner note: emphasis added. The life patterns from daily driving data are being interpreted as historical trip data] [0034]);
receiving initial data corresponding to a particular individual; inputting the initial data corresponding to a particular individual into the machine learning model (Yoshikawa: “Various exemplary embodiments provide a navigation system including a controller that collects life information associated with a user” [0009] … “The life information processing unit 26 can extract a life pattern from the life information and can register the extracted life pattern. The extracted life pattern may be stored, for example, in the life pattern memory 14. The prediction unit 27 can determine, for example, a predicted start time and a predicted destination on the basis of the life pattern information” [*Examiner note: i.e., the prediction unit 27 is being interpreted as a machine learning model] [0038] …. “First, the information providing server 11 waits for life information to be uploaded from the navigation apparatus 31 or the information terminal 32 (Step S31). … the information providing server 11 extracts a life pattern, by example, performing statistical processing on the received life information (Step S32)” [0058]),
wherein inputting the initial data corresponding to the particular individual into the machine learning model causes output of a predicted start time of a driving trip of the particular individual based at least in part the prior location data (Yoshikawa: “The life information processing unit 26 can extract a life pattern from the life information and can register the extracted life pattern. The extracted life pattern may be stored, for example, in the life pattern memory 14. The prediction unit 27 can determine, for example, a predicted start time and a predicted destination on the basis of the life pattern information” [*Examiner note: emphasis added i.e., the prediction unit 27 is being interpreted as a machine learning model] [0038]);
sending, to a mobile device corresponding to the particular individual one or more commands directing the mobile device to wake up prior to the predicted trip start time and to initiate collection of driving trip data corresponding to the driving trip (Yoshikawa: “FIG. 1 shows a navigation apparatus 31 that may be used by a user (i.e., a mobile device) … The user may be, for example, a driver or a passenger of a vehicle such as a car, a truck, a bus, or a motorcycle” [0022] … “The navigation apparatus 31 can include a controller, such as, for example, a CPU or an MPU, a storage device … a display … an input device such as a keyboard … or a touch panel” [0023] … “If it is determined that it is X minutes before the predicted start time (step S42=YES), the information providing server 11 determines whether the vehicle of the user is parked within an area in which communication with the access point is possible” [0065] … “if it is determined that the vehicle of the user is parked within an area in which communication with the access point is possible (step S42=YES), the information providing server 11 transmits a start command or signal to the navigation apparatus 31 … In response to the start command, the navigation apparatus is turned on, and thus it becomes possible for the navigation apparatus to receive data. The information providing server 11 transmits the predicted congestion information to the navigation apparatus 31 via the access point (step S45)” [*Examiner note: emphasis added] [0066]),
wherein directing the mobile device to wake up causes the mobile device to display a graphical user interface indicating that the mobile device is awake (Yoshikawa: “Therefore, data is not received from the server and is not displayed on a display means to a user until the navigation apparatus is turned on (i.e., when the device is turned on, the data is displayed on the display means).” [0007] … After a route is detected, the navigation apparatus 31 prompts the user to determine whether to download the predicted congestion information from the information providing server 11 (step S66)” [0071][FIG. 8]).
receiving the driving trip data from the mobile device input the driving trip data into the machine learning model (Yoshikawa: “Now the process performed by the information providing server 11 to extract a life pattern from the uploaded life information is described with reference to FIG. 5. FIG. 5 is a flow chart showing a process of extracting a life pattern according to an embodiment of the present invention. First, the information providing server 11 waits for life information to be uploaded from the navigation apparatus 31 or the information terminal 32 (step S31). If the information providing server 11 determines that life information has been received (step S31=YES), the information providing server 11 extracts a life pattern by, for example, performing statistical processing on the received life information (step S32)” [0058][FIG. 5]),
wherein inputting the driving trip data into the machine learning model causes output of a predicted trip end time of a driving trip of the particular individual (Yoshikawa: “The life information processing unit 26 can extract a life pattern from the life information and can register the extracted life pattern. The extracted life pattern may be stored, for example, in the life pattern memory 14. The prediction unit 27 can determine, for example, a predicted start time and a predicted destination on the basis of the life pattern information” [*Examiner note: i.e., the prediction unit 27 is being interpreted as a machine learning model] [0038] … “Additionally, a predicted arrival time may be used instead of the predicted start time, and a start point may be used instead of the destination. According to various exemplary embodiments, both the predicted start time and the predicted arrival time may be used” [0064]);
to conserve power (Yoshikawa, ¶[0024]: “In this case, in order to minimize the power consumption, it may be desirable to activate only a part of the navigation apparatus 31 necessary for receiving data.”)
Yoshikawa fails to teach:
training a machine learning model using the historical data corresponding to the driving trip patterns,
receive initial data corresponding to a particular individual, wherein the initial data includes prior location data indicating a location of the particular individual prior to entering a vehicle
communicate, to a mobile device corresponding to the particular individual and prior to the particular individual entering the vehicle, an indication of the predicted trip start time corresponding to a time proximate to the particular individual entering the vehicle, a likelihood of trip start score indicative of a likelihood that a trip will start by the predicted trip start time
one or more commands that cause the mobile device to wake up prior to the predicted trip start time and to remain awake for a predetermined amount of time associated with the likelihood of trip start score to facilitate collection, by the mobile device, of driving trip data corresponding to the driving trip,
wherein when driving trip data is not received by the mobile device within the predetermined amount of time, the mobile device transitions to a sleep state to conserve power, and
wherein a length of the predetermined amount of time increases with an increase in the likelihood that the trip will start by the predicted trip start time
communicate, to the mobile device, one or more commands directing the mobile device to stop collection of the driving trip data at the predicted trip end time, wherein sending the one or more commands directing the mobile device to stop collection of the driving trip data causes the mobile device to stop collection of the driving trip data at the predicted trip end time.
However, Lehmann teaches:
training a machine learning model using the historical data corresponding to the driving trip patterns (Lehmann: “the data storage comprises a trip history, which is a data structure comprising a multiple trip data objects. Each trip data object is a data object representing a trip that has been executed in the past” [0011] … “The trip history is used by a software component in the following referred to as a ‘learning module’ … the learning model, by training or re-training an existing machine learning algorithm on the trip history” [0014] … “According to a preferred embodiment of the invention, the default operation mode of the navigation device is to predict the destination after every start of the vehicle and to learn by adding new trip data objects to the trip history after each trip and to retrain the machine learning algorithm on a regular basis” [0028] … “In step 503, a mean squared error signal being indicative of the accuracy of the destination prediction is calculated. The weights of the starting parameters are adapted in each layer of the network by the backpropagation algorithm to minimize a mean squared error value. The mean squared error signal value is propagated backward through the network, thereby changing the weights in each layer in a way minimizing the error value … The re-trained and improved version of the destination prediction algorithm 207 is returned in step 504 from the learning module 206 to the destination prediction module 208” [0103][FIG. 5]);
wherein when driving trip data is not received by the mobile device within the predetermined amount of time, the mobile device transitions to a sleep state to conserve power, and (Lehmann, ¶[0044]: “According to preferred embodiments of the invention, at least three operation modes of the navigation device are supported: an 'instructed driving mode', a 'commuter mode and a 'silent mode' and at least a first and a second threshold variable for the prediction score are specified by a user or manufacturer of the navigation device.”, ¶[0051]: “According to some embodiments of the invention, a second prediction score threshold variable is in the range between 0 and 50%. In case the prediction score of all predicted destinations is below said second threshold, the navigation device switches to 'silent mode' and does not provide the driver with instructions on where to drive to or present him predicted destinations.”, ¶[0052]: “In 'silent mode', the navigation device remains silent until, at some point along the route actually chosen by the driver, at least one destination can be predicted with a prediction score higher than the first or the second prediction score threshold variable. In this case, the navigation device will switch to another operation mode.”) [Examiner’s note: “driving trip data is not received by the mobile device within the predetermined amount of time” e.g., “the navigation device does not provide the driver with instructions on where to drive or present him predicted destinations”]
wherein a length of the predetermined amount of time increases with an increase in the likelihood that the trip will start by the predicted trip start time (Lehmann, ¶[0045]: “The 'commuter mode' is a mode of operation wherein the navigation device has predicted at least one destination with high prediction score (high probability value for a particular destination given a high or at least a sufficient prediction algorithm accuracy) and wherein said prediction score lies above a first threshold variable. According to some embodiments of the invention, said first prediction score threshold variable is in the range of 70- 100%. The 'commuter mode' is a mode of operation of the navigation device according to which all those functions of the navigation device are enabled which provide a beneficial effect for a user who is familiar with a particular route.”, ¶[0047]: “In case the calculated prediction score for all predicted destinations is lower than the first threshold variable, but at least one predicted destination has a prediction score being higher than a second prediction score threshold variable, the navigation device is operated in an 'instructed driving mode'. In this mode, functions which require a high reliability of the predictions are disabled or at least not executed automatically.”, ¶[0051]: “According to some embodiments of the invention, a second prediction score threshold variable is in the range between 0 and 50%. In case the prediction score of all predicted destinations is below said second threshold, the navigation device switches to 'silent mode' and does not provide the driver with instructions on where to drive to or present him predicted destinations.”) [Examiner’s note: The highlight indicates that if the device determines that the likelihood of trip prediction score exceeds a first predetermined threshold (e.g., 70 – 100%), the mobile device remains in a “commuter mode” (e.g., remains awake) with all the functions of the navigation device enabled and if the mobile device determines that the likelihood of prediction score does not exceed the first predetermined threshold (e.g., a second prediction score threshold is in the range between 0-50%), the navigation device switches to a “instructed driving mode” with some functions are disabled (e.g., the device remains awake for a shorter period of time than the first period of time)]
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa and Lehmann. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data. Lehmann discloses training a machine learning model on driver data. One of ordinary skill would have motivation to combine Yoshikawa and Lehmann to train the statistical processing model taught by Yoshikawa with the machine learning training taught by Lehmann because utilizing a machine learning model trained on trip history improves the accuracy of the destination prediction. (Lehmann: “The learning module, by training or retraining an existing machine learning algorithm on the trip history, generates a destination prediction algorithm whose prediction accuracy improves over time” [0015]).
However, Loriaux explicitly discloses:
receive initial data corresponding to a particular individual, wherein the initial data includes prior location data indicating a location of the particular individual prior to entering a vehicle (Loriaux, ¶[0008]: “When executed by the one or more processors, the instructions cause the computing system to receive user data indicative of (i) a plurality of locations of a user while the user was driving and (ii) times at which the user was driving at the plurality of locations, and store the received user data in the database. The instructions also cause the computing system to provide, at a notification time that is at or prior to the time when the user is likely to begin driving the route, a notification to a mobile device of the user, the notification indicating one or both of (i) the one or more entities and (ii) locations of the one or more entities.”)
communicate, to a mobile device corresponding to the particular individual and prior to the particular individual entering the vehicle, an indication of the predicted trip start time corresponding to a time proximate to the particular individual entering the vehicle, (Loriaux, ¶[0008]: “The instructions also cause the computing system to determine, by analyzing the user data stored in the database, a route that the user is likely to drive and a time when the user is likely to begin driving the route. The instructions also cause the computing system to identify one or more entities each having a location proximate to the route, at least in part by analyzing, for each of the one or more entities, (i) a current price of a good or service provided by the entity and (ii) current prices of goods or services provided by one or more other entities. The instructions also cause the computing system to provide, at a notification time that is at or prior to the time when the user is likely to begin driving the route, a notification to a mobile device of the user, the notification indicating one or both of (i) the one or more entities and (ii) locations of the one or more entities.”)
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa and Loriaux. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. One of ordinary skill would have motivation to combine Yoshikawa and Loriaux to enable the individual to coordinate their arrival at the vehicle more efficiently, reducing idle time and unnecessary waiting. This early notification can prompt the individual to complete pre-departure tasks, prepare personal items, or navigate toward the vehicle in a timely manner. Additionally, aligning the individual’s arrival with the predicted trip start helps optimize vehicle readiness functions, thereby improving energy usage and overall user experience.
However, Spears explicitly discloses:
communicate, to a mobile device corresponding to the particular individual, an
indication of the predicted trip start time, a likelihood of trip start score indicative of a
likelihood that a trip will start by the predicted trip start time, and (Spears, ,i [0033]:
"In FIG. 1, the server (501) predicts the time period and the route of the next travel Qf.
the mobile device ( 405) based on the route that is recently traversed by the mobile
device (405), uses the prediction to select location-based information (503), and
schedules the location-based information (503) for delivery to the mobile device ( 405)
a predetermined time period before the mobile device (405) starts to travel along the
predicted route in accordance with the predicted time period of travel. Thus, the user of
the mobile device ( 405) is provided with an optimal time window to review the
information and arrange the travel using the information.", ,i[0042]: "The predicted
route and time of the trip can be used to select and/or present offers along or near the
route, based on utility scores of the offers that are a function of the closet to the route
as well as most likely to be redeemed.", ,i[0052]: "In FIG. 2, the route dictionary (232)
includes a list of routes ( 401, ... , 403) of past trips of a user (101 ), the frequencies (411,
... , 413) of the user (101) traveling along the routes (401, ... , 403), and the times traveled (421, ... , 423) (e.g., time windows within a day, week, and/or month).")
one or more commands that cause the mobile device to wake up prior to the predicted trip start time and to remain awake for a predetermined amount of time associated with the likelihood of trip start score to facilitate collection, by the mobile device, of driving trip data corresponding to the driving trip, (Spears, ¶[0033]: “the server (501) predicts the time period and the route of the next travel of the mobile device ( 405) based on the route that is recently traversed by the mobile device (405), uses the prediction to select location-based information (503), and schedules the location-based information (503) for delivery to the mobile device ( 405) a predetermined time period before the mobile device (405) starts to travel along the predicted route in accordance with the predicted time period of travel. Thus, the user of the mobile device ( 405) is provided with an optimal time window to review the information and arrange the travel using the information.”, ¶[0040]: “For example, route candidates can be selected from the route dictionary (232) based on whether the time periods of the route are within a predetermined time window (e.g., next hour)”, ¶[0045]: “One or more top ranked offer candidates can be selected for presentation to the user at a predetermined time prior to the next trip.”) [Examiner’s note: The highlight indicates that the mobile device stays awake for a predetermined amount of time associated with the trip start time. The fact that one or more top ranked route selection candidates being presented to the user via the mobile device at a predetermined time prior to the next trip indicates that the mobile device remains awake during that time]
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa and Spears. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. One of ordinary skill would have motivation to combine Yoshikawa and Spears to ensure the mobile device responsible for data collection is active at the right time while optimizing power consumption. By associating the likelihood of a trip start with a dynamically adjustable “wake-up” period, this approach minimizes unnecessary device activity, extending battery life and reducing energy waste.
However, Makoto discloses:
communicate, to the mobile device, one or more commands directing the mobile device to stop collection of the driving trip data at the predicted trip end time, wherein sending the one or more commands directing the mobile device to stop collection of the driving trip data causes the mobile device to stop collection of the driving trip data at the predicted trip end time (Makoto: “In the interval driving data storing unit 101, driving data of a vehicle, which is a collection of sampling data sampled at predetermined period (for example, 200 msec) in certain driving interval, is stored. The driving interval has certain time range (for example, 1 minute or 10 seconds etc.) from start point time to end point time.” [*Examiner note: emphasis added. i.e., the start time and end time of the interval would be the predicted start time and the predicted end time of the Yoshikawa reference, and collecting data during an interval would result in data collection stopping at the end of the interval (i.e., at the predicted end time)] [0029]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Lehmann and Makoto. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data and Lehmann discloses training a machine learning model on driver data. Makoto discloses collecting data over an interval from a start time to an end time. One of ordinary skill would have motivation to combine Yoshikawa in view of Lehmann and Makoto to replace the end time of the interval for data collection taught by Makoto with the predicted trip end time disclosed by Yoshikawa in view of Lehmann because useful driver information can be determined by only gathering data during an interval of time (Makoto: “It is therefore important to evaluate a driving result in a time interval from a past relative short time to the present time point by using only interval driving data as driving data in the interval and immediately provide the evaluation result to the driver.” [0006]).
Regarding claim 11, Yoshikawa in view of Lehmann, Spears, Loriaux and in further view of Makoto discloses all of the limitations of claim 1 as shown in the rejection above. Lehmann also discloses:
wherein receiving the historical data comprises receiving one or more of: global positioning system (GPS) data, time information, demographics information, income information, accelerometer data, gyroscope data, barometer data, magnetometer data, or social media data (Lehmann: Starting parameters are gathered in steps 301-304 as described previously … A type of starting parameter could be, for example, “current outdoor temperature”, “time”, “day in week” (i.e., time information) [0102]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Makoto, Spears, Loriaux and Lehmann. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data, and Makoto discloses collecting data over an interval from a start time to an end time. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. Lehmann discloses training a machine learning model on driver data. One of ordinary skill would have motivation to combine Yoshikawa in view of Makoto, Spears, Loriaux and Lehmann to include time information in the historical data as taught by Lehmann in the training data disclosed by Yoshikawa in view of Makoto and Spears because this additional time information improves the accuracy of the machine learning model’s predictions (Lehmann: “The accuracy of the destination prediction algorithm can be further refined by taking into consideration in addition the date and date related information. For example, the destination chosen by a driver may strongly depend on the question, if the day is a working day, Saturday or Sunday or Holiday” [0032]).
Regarding claim 13, Yoshikawa in view of Lehmann, Spears, Loriaux and in further view of Makoto discloses all of the limitations of claim 10 as shown in the rejection above. Lehmann also discloses:
wherein the historical data is labelled based on a corresponding historical driving trips (Lehmann: The trip history is a data structure, e.g. a data object of an object-oriented programming language or a table in a relational data base or any other application structure operable to store a set of trip data objects 201 representing past trips. Each trip data object comprises at least a destination information 205 and a set of (i.e., are labeled by) starting parameters 204, wherein the set of starting parameters comprise at least the actual destination of a trip, its starting location, and its starting time and date” [0074]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Makoto, Spears, Loriaux and Lehmann. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data, and Makoto discloses collecting data over an interval from a start time to an end time. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. Lehmann discloses training a machine learning model on driver data. One of ordinary skill would have motivation to combine Yoshikawa in view of Makoto, Spears, Loriaux and Lehmann to include time information in the historical data as taught by Lehmann in the training data disclosed by Yoshikawa in view of Makoto and Spears because this additional time information improves the accuracy of the machine learning model’s predictions (Lehmann: “The accuracy of the destination prediction algorithm can be further refined by taking into consideration in addition the date and date related information. For example, the destination chosen by a driver may strongly depend on the question, if the day is a working day, Saturday or Sunday or Holiday” [0032]).
Regarding claim 15, Yoshikawa in view of Lehmann, Spears, Loriaux and in further view of Makoto discloses all of the limitations of claim 10 as shown in the rejection above. Lehmann also discloses:
wherein the mobile device is not configured to collect driving trip data prior to waking up (Yoshikawa: “communication between a navigation apparatus and a server only starts after the navigation apparatus is turned on. Therefore, data is not received from the server and is not displayed on a display means to a user until the navigation apparatus is turned on” [0007]).
Regarding claim 17, Yoshikawa in view of Lehmann, Spears, Loriaux and in further view of Makoto discloses all of the limitations of claim 10 as shown in the rejection above. Lehmann also discloses:
further comprising: dynamically updating the machine learning model based on the driving trip data (Lehmann: “The weights of the starting parameters are adapted in each layer of the network by the back-propagation algorithm to minimize a mean squared error value. The mean squared error signal value is propagated backward through the network, thereby changing the weights in each layer in a way minimizing the error value. Provided the driver shows a steady driving behavior regarding his chosen destination, the prediction accuracy will increase upon every added new trip data object and every re-training of the neural network on the increased set of trip data objects” [0060][Figure. 5 step 503]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Makoto, Spears, Loriaux and Lehmann. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data, and Makoto discloses collecting data over an interval from a start time to an end time. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. Lehmann discloses training a machine learning model on driver data. One of ordinary skill would have motivation to combine Yoshikawa in view of Makoto, Spears, Loriaux and Lehmann to continuously update the model as taught by Lehmann using the training data disclosed by Yoshikawa in view of Makoto, Loriaux and Spears because updating the machine learning model trained on trip history improves the accuracy of the destination prediction. (Lehmann: “The learning module, by training or retraining an existing machine learning algorithm on the trip history, generates a destination prediction algorithm whose prediction accuracy improves over time” [0015]).
Regarding claim 18, Yoshikawa teaches:
One or more non-transitory computer-readable media storing instructions that, when executed by a computing platform comprising at least one processor, a communication interface, and memory, cause the computing platform to receive historical data corresponding to driving trip patterns (Yoshikawa: “FIG. 1 is a conceptual diagram of a navigation apparatus according to an exemplary embodiment of the invention. FIG. 1 shows an information providing server 11 implemented, for example, in a computer that may include a controller such as, for example, a CPU or an MPU storage means such as a semiconductor memory, a magnetic disk, or an optical disk and a communication interface” [0021] [Claim 19: “A storage medium storing a set of program instructions executable on a data processing device”] … “The life pattern memory 14 (e.g., a database) stores life pattern data produced by extracting life patterns from daily driving data and/or life information associated with the user, such as schedule information, received from the navigation apparatus 31 or the information terminal 32 of the user.” [*Examiner note: emphasis added. The life patterns from daily driving data are being interpreted as historical trip data] [0034]);
receive initial data corresponding to a particular individual; input the initial data corresponding to a particular individual into the machine learning model (Yoshikawa: “Various exemplary embodiments provide a navigation system including a controller that collects life information associated with a user” [0009] … “The life information processing unit 26 can extract a life pattern from the life information and can register the extracted life pattern. The extracted life pattern may be stored, for example, in the life pattern memory 14. The prediction unit 27 can determine, for example, a predicted start time and a predicted destination on the basis of the life pattern information” [*Examiner note: i.e., the prediction unit 27 is being interpreted as a machine learning model] [0038] …. “First, the information providing server 11 waits for life information to be uploaded from the navigation apparatus 31 or the information terminal 32 (Step S31). … the information providing server 11 extracts a life pattern, by example, performing statistical processing on the received life information (Step S32)” [0058]),
wherein inputting the initial data corresponding to the particular individual into the machine learning model causes output of a predicted start time of a driving trip of the particular individual based at least in part the prior location data (Yoshikawa: “The life information processing unit 26 can extract a life pattern from the life information and can register the extracted life pattern. The extracted life pattern may be stored, for example, in the life pattern memory 14. The prediction unit 27 can determine, for example, a predicted start time and a predicted destination on the basis of the life pattern information” [*Examiner note: emphasis added i.e., the prediction unit 27 is being interpreted as a machine learning model] [0038]);
send, prior to the predicted start time and to a mobile device corresponding to the particular individual, one or more commands directing the mobile device to wake up prior to the predicted start time and to initiate collection of driving trip data corresponding to the driving trip, wherein directing the mobile device to wake up causes the mobile device to be configured for the collection of driving trip data (Yoshikawa: “FIG. 1 shows a navigation apparatus 31 that may be used by a user (i.e., a mobile device) … The user may be, for example, a driver or a passenger of a vehicle such as a car, a truck, a bus, or a motorcycle” [0022] … “The navigation apparatus 31 can include a controller, such as, for example, a CPU or an MPU, a storage device … a display … an input device such as a keyboard … or a touch panel” [0023] … “If it is determined that it is X minutes before the predicted start time (step S42=YES), the information providing server 11 determines whether the vehicle of the user is parked within an area in which communication with the access point is possible” [0065] … “if it is determined that the vehicle of the user is parked within an area in which communication with the access point is possible (step S42=YES), the information providing server 11 transmits a start command or signal to the navigation apparatus 31 … In response to the start command, the navigation apparatus is turned on, and thus it becomes possible for the navigation apparatus to receive data. The information providing server 11 transmits the predicted congestion information to the navigation apparatus 31 via the access point (step S45)” [*Examiner note: i.e., emphasis added] [0066]).
receive the driving trip data from the mobile device input the driving trip data into the machine learning model (Yoshikawa: “Now the process performed by the information providing server 11 to extract a life pattern from the uploaded life information is described with reference to FIG. 5. FIG. 5 is a flow chart showing a process of extracting a life pattern according to an embodiment of the present invention. First, the information providing server 11 waits for life information to be uploaded from the navigation apparatus 31 or the information terminal 32 (step S31). If the information providing server 11 determines that life information has been received (step S31=YES), the information providing server 11 extracts a life pattern by, for example, performing statistical processing on the received life information (step S32)” [0058][FIG. 5]),
wherein inputting the driving trip data into the machine learning model causes output of a predicted trip end time of a driving trip of the particular individual based at least in part the prior location data (Yoshikawa: “The life information processing unit 26 can extract a life pattern from the life information and can register the extracted life pattern. The extracted life pattern may be stored, for example, in the life pattern memory 14. The prediction unit 27 can determine, for example, a predicted start time and a predicted destination on the basis of the life pattern information” [*Examiner note: i.e., the prediction unit 27 is being interpreted as a machine learning model] [0038] … “Additionally, a predicted arrival time may be used instead of the predicted start time, and a start point may be used instead of the destination. According to various exemplary embodiments, both the predicted start time and the predicted arrival time may be used” [0064]);
to conserve power (Yoshikawa, ¶[0024]: “In this case, in order to minimize the power consumption, it may be desirable to activate only a part of the navigation apparatus 31 necessary for receiving data.”)
Yoshikawa fails to teach:
train a machine learning model using the historical data corresponding to the driving trip patterns,
receive initial data corresponding to a particular individual, wherein the initial data includes prior location data indicating a location of the particular individual prior to entering a vehicle
communicate, to a mobile device corresponding to the particular individual and prior to the particular individual entering the vehicle, an indication of the predicted trip start time corresponding to a time proximate to the particular individual entering the vehicle, a likelihood of trip start score indicative of a likelihood that a trip will start by the predicted trip start time,
one or more commands that cause the mobile device to wake up prior to the predicted trip start time and to remain awake for a predetermined amount of time associated with the likelihood of trip start score to facilitate collection, by the mobile device, of driving trip data corresponding to the driving trip
wherein when driving trip data is not received by the mobile device within the predetermined amount of time, the mobile device transitions to a sleep state to conserve power, and
wherein a length of the predetermined amount of time increases with an increase in the likelihood that the trip will start by the predicted trip start time
communicate, to the mobile device, one or more commands directing the mobile device to stop collection of the driving trip data at the predicted trip end time, wherein sending the one or more commands directing the mobile device to stop collection of the driving trip data causes the mobile device to stop collection of the driving trip data at the predicted trip end time.
However, Lehmann teaches:
train a machine learning model using the historical data corresponding to the driving trip patterns (Lehmann: “the data storage comprises a trip history, which is a data structure comprising a multiple trip data objects. Each trip data object is a data object representing a trip that has been executed in the past” [0011] … “The trip history is used by a software component in the following referred to as a ‘learning module’ … the learning model, by training or re-training an existing machine learning algorithm on the trip history” [0014] … “According to a preferred embodiment of the invention, the default operation mode of the navigation device is to predict the destination after every start of the vehicle and to learn by adding new trip data objects to the trip history after each trip and to retrain the machine learning algorithm on a regular basis” [0028] … “In step 503, a mean squared error signal being indicative of the accuracy of the destination prediction is calculated. The weights of the starting parameters are adapted in each layer of the network by the backpropagation algorithm to minimize a mean squared error value. The mean squared error signal value is propagated backward through the network, thereby changing the weights in each layer in a way minimizing the error value … The re-trained and improved version of the destination prediction algorithm 207 is returned in step 504 from the learning module 206 to the destination prediction module 208” [0103][FIG. 5]);
wherein when driving trip data is not received by the mobile device within the predetermined amount of time, the mobile device transitions to a sleep state to conserve power, and (Lehmann, ¶[0044]: “According to preferred embodiments of the invention, at least three operation modes of the navigation device are supported: an 'instructed driving mode', a 'commuter mode and a 'silent mode' and at least a first and a second threshold variable for the prediction score are specified by a user or manufacturer of the navigation device.”, ¶[0051]: “According to some embodiments of the invention, a second prediction score threshold variable is in the range between 0 and 50%. In case the prediction score of all predicted destinations is below said second threshold, the navigation device switches to 'silent mode' and does not provide the driver with instructions on where to drive to or present him predicted destinations.”, ¶[0052]: “In 'silent mode', the navigation device remains silent until, at some point along the route actually chosen by the driver, at least one destination can be predicted with a prediction score higher than the first or the second prediction score threshold variable. In this case, the navigation device will switch to another operation mode.”) [Examiner’s note: “driving trip data is not received by the mobile device within the predetermined amount of time” e.g., “the navigation device does not provide the driver with instructions on where to drive or present him predicted destinations”]
wherein a length of the predetermined amount of time increases with an increase in the likelihood that the trip will start by the predicted trip start time (Lehmann, ¶[0045]: “The 'commuter mode' is a mode of operation wherein the navigation device has predicted at least one destination with high prediction score (high probability value for a particular destination given a high or at least a sufficient prediction algorithm accuracy) and wherein said prediction score lies above a first threshold variable. According to some embodiments of the invention, said first prediction score threshold variable is in the range of 70- 100%. The 'commuter mode' is a mode of operation of the navigation device according to which all those functions of the navigation device are enabled which provide a beneficial effect for a user who is familiar with a particular route.”, ¶[0047]: “In case the calculated prediction score for all predicted destinations is lower than the first threshold variable, but at least one predicted destination has a prediction score being higher than a second prediction score threshold variable, the navigation device is operated in an 'instructed driving mode'. In this mode, functions which require a high reliability of the predictions are disabled or at least not executed automatically.”, ¶[0051]: “According to some embodiments of the invention, a second prediction score threshold variable is in the range between 0 and 50%. In case the prediction score of all predicted destinations is below said second threshold, the navigation device switches to 'silent mode' and does not provide the driver with instructions on where to drive to or present him predicted destinations.”) [Examiner’s note: The highlight indicates that if the device determines that the likelihood of trip prediction score exceeds a first predetermined threshold (e.g., 70 – 100%), the mobile device remains in a “commuter mode” (e.g., remains awake) with all the functions of the navigation device enabled and if the mobile device determines that the likelihood of prediction score does not exceed the first predetermined threshold (e.g., a second prediction score threshold is in the range between 0-50%), the navigation device switches to a “instructed driving mode” with some functions are disabled (e.g., the device remains awake for a shorter period of time than the first period of time)]
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa and Lehmann. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data. Lehmann discloses training a machine learning model on driver data. One of ordinary skill would have motivation to combine Yoshikawa and Lehmann to train the statistical processing model taught by Yoshikawa with the machine learning training taught by Lehmann because utilizing a machine learning model trained on trip history improves the accuracy of the destination prediction. (Lehmann: “The learning module, by training or retraining an existing machine learning algorithm on the trip history, generates a destination prediction algorithm whose prediction accuracy improves over time” [0015]).
However, Loriaux explicitly discloses:
receive initial data corresponding to a particular individual, wherein the initial data includes prior location data indicating a location of the particular individual prior to entering a vehicle (Loriaux, ¶[0008]: “When executed by the one or more processors, the instructions cause the computing system to receive user data indicative of (i) a plurality of locations of a user while the user was driving and (ii) times at which the user was driving at the plurality of locations, and store the received user data in the database. The instructions also cause the computing system to provide, at a notification time that is at or prior to the time when the user is likely to begin driving the route, a notification to a mobile device of the user, the notification indicating one or both of (i) the one or more entities and (ii) locations of the one or more entities.”)
communicate, to a mobile device corresponding to the particular individual and prior to the particular individual entering the vehicle, an indication of the predicted trip start time corresponding to a time proximate to the particular individual entering the vehicle, (Loriaux, ¶[0008]: “The instructions also cause the computing system to determine, by analyzing the user data stored in the database, a route that the user is likely to drive and a time when the user is likely to begin driving the route. The instructions also cause the computing system to identify one or more entities each having a location proximate to the route, at least in part by analyzing, for each of the one or more entities, (i) a current price of a good or service provided by the entity and (ii) current prices of goods or services provided by one or more other entities. The instructions also cause the computing system to provide, at a notification time that is at or prior to the time when the user is likely to begin driving the route, a notification to a mobile device of the user, the notification indicating one or both of (i) the one or more entities and (ii) locations of the one or more entities.”)
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa and Loriaux. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. One of ordinary skill would have motivation to combine Yoshikawa and Loriaux to enable the individual to coordinate their arrival at the vehicle more efficiently, reducing idle time and unnecessary waiting. This early notification can prompt the individual to complete pre-departure tasks, prepare personal items, or navigate toward the vehicle in a timely manner. Additionally, aligning the individual’s arrival with the predicted trip start helps optimize vehicle readiness functions, thereby improving energy usage and overall user experience.
However, Spears explicitly discloses:
communicate, to a mobile device corresponding to the particular individual, an
indication of the predicted trip start time, a likelihood of trip start score indicative of a
likelihood that a trip will start by the predicted trip start time, and (Spears, ,i [0033]:
"In FIG. 1, the server (501) predicts the time period and the route of the next travel Qf.
the mobile device ( 405) based on the route that is recently traversed by the mobile
device (405), uses the prediction to select location-based information (503), and
schedules the location-based information (503) for delivery to the mobile device ( 405)
a predetermined time period before the mobile device (405) starts to travel along the
predicted route in accordance with the predicted time period of travel. Thus, the user of
the mobile device ( 405) is provided with an optimal time window to review the
information and arrange the travel using the information.", ,i[0042]: "The predicted
route and time of the trip can be used to select and/or present offers along or near the
route, based on utility scores of the offers that are a function of the closet to the route
as well as most likely to be redeemed.", ,i[0052]: "In FIG. 2, the route dictionary (232)
includes a list of routes ( 401, ... , 403) of past trips of a user (101 ), the frequencies (411,
... , 413) of the user (101) traveling along the routes (401, ... , 403), and the times traveled (421, ... , 423) (e.g., time windows within a day, week, and/or month).")
one or more commands that cause the mobile device to wake up prior to the predicted trip start time and to remain awake for a predetermined amount of time associated with the likelihood of trip start score to facilitate collection, by the mobile device, of driving trip data corresponding to the driving trip, (Spears, ¶[0033]: “the server (501) predicts the time period and the route of the next travel of the mobile device ( 405) based on the route that is recently traversed by the mobile device (405), uses the prediction to select location-based information (503), and schedules the location-based information (503) for delivery to the mobile device ( 405) a predetermined time period before the mobile device (405) starts to travel along the predicted route in accordance with the predicted time period of travel. Thus, the user of the mobile device ( 405) is provided with an optimal time window to review the information and arrange the travel using the information.”, ¶[0040]: “For example, route candidates can be selected from the route dictionary (232) based on whether the time periods of the route are within a predetermined time window (e.g., next hour)”, ¶[0045]: “One or more top ranked offer candidates can be selected for presentation to the user at a predetermined time prior to the next trip.”) [Examiner’s note: The highlight indicates that the mobile device stays awake for a predetermined amount of time associated with the trip start time. The fact that one or more top ranked route selection candidates being presented to the user via the mobile device at a predetermined time prior to the next trip indicates that the mobile device remains awake during that time]
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa and Spears. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. One of ordinary skill would have motivation to combine Yoshikawa and Spears to ensure the mobile device responsible for data collection is active at the right time while optimizing power consumption. By associating the likelihood of a trip start with a dynamically adjustable “wake-up” period, this approach minimizes unnecessary device activity, extending battery life and reducing energy waste.
However, Makoto discloses:
communicate, to the mobile device, one or more commands directing the mobile device to stop collection of the driving trip data at the predicted trip end time, wherein sending the one or more commands directing the mobile device to stop collection of the driving trip data causes the mobile device to stop collection of the driving trip data at the predicted trip end time (Makoto: “In the interval driving data storing unit 101, driving data of a vehicle, which is a collection of sampling data sampled at predetermined period (for example, 200 msec) in certain driving interval, is stored. The driving interval has certain time range (for example, 1 minute or 10 seconds etc.) from start point time to end point time.” [*Examiner note: emphasis added. i.e., the start time and end time of the interval would be the predicted start time and the predicted end time of the Yoshikawa reference, and collecting data during an interval would result in data collection stopping at the end of the interval (i.e., at the predicted end time)] [0029]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Lehmann and Makoto. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data and Lehmann discloses training a machine learning model on driver data. Makoto discloses collecting data over an interval from a start time to an end time. One of ordinary skill would have motivation to combine Yoshikawa in view of Lehmann and Makoto to replace the end time of the interval for data collection taught by Makoto with the predicted trip end time disclosed by Yoshikawa in view of Lehmann because useful driver information can be determined by only gathering data during an interval of time (Makoto: “It is therefore important to evaluate a driving result in a time interval from a past relative short time to the present time point by using only interval driving data as driving data in the interval and immediately provide the evaluation result to the driver.” [0006]).
Regarding claim 19, Yoshikawa in view of Lehmann, Spears, Loriaux and in further view of Makoto discloses all of the limitations of claim 18 as shown in the rejection above. Lehmann also discloses:
wherein receiving the historical data comprises receiving one or more of: global positioning system (GPS) data, time information, demographics information, income information, accelerometer data, gyroscope data, barometer data, magnetometer data, or social media data (Lehmann: Starting parameters are gathered in steps 301-304 as described previously … A type of starting parameter could be, for example, “current outdoor temperature”, “time”, “day in week” (i.e., time information) [0102]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Makoto, Spears, Loriaux and Lehmann. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data, and Makoto discloses collecting data over an interval from a start time to an end time. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. Lehmann discloses training a machine learning model on driver data. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. One of ordinary skill would have motivation to combine Yoshikawa in view of Makoto, Spears, Loriaux and Lehmann to include time information in the historical data as taught by Lehmann in the training data disclosed by Yoshikawa in view of Makoto, Loriaux and Spears because this additional time information improves the accuracy of the machine learning model’s predictions (Lehmann: “The accuracy of the destination prediction algorithm can be further refined by taking into consideration in addition the date and date related information. For example, the destination chosen by a driver may strongly depend on the question, if the day is a working day, Saturday or Sunday or Holiday” [0032]).
Claims 3, 12, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Yoshikawa in view of Lehmann in further view of Makoto, Spears, Loriaux and further in view of Anthony (US PGPUB No US2018/0315260 A1) (hereafter referred to as Anthony).
Regarding claim 3, Yoshikawa in view of Lehmann, Spears, Makoto and in further view of Loriaux discloses all of the limitations of claim 1 as shown in the rejection above. Yoshikawa in view of Lehmann, Spears, Makoto and in further view of Loriaux fails to disclose:
wherein training the machine learning model further comprises validating a first subset of the historical data with a second subset of the historical data, wherein the first subset of the historical data is received from the mobile device and the second subset of the historical data is received from an on board diagnostics (OBD) system.
However, Anthony teaches:
wherein training the machine learning model further comprises validating a first subset of the historical data with a second subset of the historical data, wherein the first subset of the historical data is received from the mobile device and the second subset of the historical data is received from an on board diagnostics (OBD) system (Anthony: “FIG. 1 illustrates how sensors (1) feed signal processing elements (2) which in turn drive decisions by a machine learning (ML) model (3). One or more sensors (1), also called “listening devices” herein, may include audio, vibration, electromagnetic, or other sensors … may be provided by the operator’s personal mobile device such as a smart phone or tablet” [0026][FIG. 1] … “Signal processing (2) is then applied to the sensor outputs … regardless of implementation, the filters/signal processing (2) tries to extract one or more features of interest out of the signals generated by the sensors” [0027] … “In one approach, a companion OBD-II read reader is plugged into the car (or otherwise accessed) to read diagnostic codes in real time while a nearby device (phone or special hardware) is also collecting sensor data. The ML model makes determinations about the diagnosis of the vehicle (healthy or specific error). That determination will be compared to the OBD-II readout (or smartphone or other device sensors) to provide feedback to the machine learning model. This will either reinforce the model if the diagnosis was correct, or tell the ML model it was incorrect and reclassify it appropriately” [*Examiner note: i.e. validating the first subset with the second subset] [0036]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Lehmann in further view of Makoto, Spears, Loriaux and Anthony. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data, Lehmann discloses training a machine learning model on driver data, and Makoto discloses collecting data over an interval from a start time to an end time. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. Anthony discloses validating driving data for machine learning. One of ordinary skill would have motivation to combine Yoshikawa in view of Lehmann in further view of Makoto, Spears, Loriaux and Anthony to validate the data collected by Yoshikawa in view of Lehmann in further view of Makoto, Loriaux and Spears using the data validation taught by Anthony because validating this data helps train the machine learning model more accurately (Anthony: “Subsets of the sensors filters, and neural networks diagnose vehicle faults or other problems by building unique machine learning models for each vehicle. Because the approach can adapt, it can provide improvements in the quality of diagnostics over time” [0006]).
Regarding claim 12, Yoshikawa in view of Lehmann, Spears, Makoto and in further view of Loriaux discloses all of the limitations of claim 10 as shown in the rejection above. Yoshikawa in view of Lehmann, Spears, Makoto and in further view of Loriaux fails to disclose:
wherein training the machine learning model further comprises validating a first subset of the historical data with a second subset of the historical data, wherein the first subset of the historical data is received from the mobile device and the second subset of the historical data is received from an on board diagnostics (OBD) system.
However, Anthony teaches:
wherein training the machine learning model further comprises validating a first subset of the historical data with a second subset of the historical data, wherein the first subset of the historical data is received from the mobile device and the second subset of the historical data is received from an on board diagnostics (OBD) system (Anthony: “FIG. 1 illustrates how sensors (1) feed signal processing elements (2) which in turn drive decisions by a machine learning (ML) model (3). One or more sensors (1), also called “listening devices” herein, may include audio, vibration, electromagnetic, or other sensors … may be provided by the operator’s personal mobile device such as a smart phone or tablet” [0026][FIG. 1] … “Signal processing (2) is then applied to the sensor outputs … regardless of implementation, the filters/signal processing (2) tries to extract one or more features of interest out of the signals generated by the sensors” [0027] … “In one approach, a companion OBD-II read reader is plugged into the car (or otherwise accessed) to read diagnostic codes in real time while a nearby device (phone or special hardware) is also collecting sensor data. The ML model makes determinations about the diagnosis of the vehicle (healthy or specific error). That determination will be compared to the OBD-II readout (or smartphone or other device sensors) to provide feedback to the machine learning model. This will either reinforce the model if the diagnosis was correct, or tell the ML model it was incorrect and reclassify it appropriately” [*Examiner note: i.e. validating the first subset with the second subset] [0036]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Lehmann, Spears in further view of Makoto, Loriaux and Anthony. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data, Lehmann discloses training a machine learning model on driver data, and Makoto discloses collecting data over an interval from a start time to an end time. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. Anthony discloses validating driving data for machine learning. One of ordinary skill would have motivation to combine Yoshikawa in view of Lehmann in further view of Makoto, Spears, Loriaux and Anthony to validate the data collected by Yoshikawa in view of Lehmann in further view of Makoto, Loriaux and Spears using the data validation taught by Anthony because validating this data helps train the machine learning model more accurately (Anthony: “Subsets of the sensors filters, and neural networks diagnose vehicle faults or other problems by building unique machine learning models for each vehicle. Because the approach can adapt, it can provide improvements in the quality of diagnostics over time” [0006]).
Regarding claim 20, Yoshikawa in view of Lehmann, Spears, Loriaux and in further view of Makoto discloses all of the limitations of claim 18 as shown in the rejection above. Yoshikawa in view of Lehmann, Spears, Makoto and in further view of Loriaux fails to disclose:
wherein training the machine learning model further comprises validating a first subset of the historical data with a second subset of the historical data, wherein the first subset of the historical data is received from the mobile device and the second subset of the historical data is received from an on board diagnostics (OBD) system.
However, Anthony teaches:
wherein training the machine learning model further comprises validating a first subset of the historical data with a second subset of the historical data, wherein the first subset of the historical data is received from the mobile device and the second subset of the historical data is received from an on board diagnostics (OBD) system (Anthony: “FIG. 1 illustrates how sensors (1) feed signal processing elements (2) which in turn drive decisions by a machine learning (ML) model (3). One or more sensors (1), also called “listening devices” herein, may include audio, vibration, electromagnetic, or other sensors … may be provided by the operator’s personal mobile device such as a smart phone or tablet” [0026][FIG. 1] … “Signal processing (2) is then applied to the sensor outputs … regardless of implementation, the filters/signal processing (2) tries to extract one or more features of interest out of the signals generated by the sensors” [0027] … “In one approach, a companion OBD-II read reader is plugged into the car (or otherwise accessed) to read diagnostic codes in real time while a nearby device (phone or special hardware) is also collecting sensor data. The ML model makes determinations about the diagnosis of the vehicle (healthy or specific error). That determination will be compared to the OBD-II readout (or smartphone or other device sensors) to provide feedback to the machine learning model. This will either reinforce the model if the diagnosis was correct, or tell the ML model it was incorrect and reclassify it appropriately” [*Examiner note: i.e. validating the first subset with the second subset] [0036]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Lehmann, Spears in further view of Makoto, Loriaux and Anthony. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data, Lehmann discloses training a machine learning model on driver data, and Makoto discloses collecting data over an interval from a start time to an end time. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. Anthony discloses validating driving data for machine learning. One of ordinary skill would have motivation to combine Yoshikawa in view of Lehmann, Spears in further view of Makoto, Loriaux and Anthony to validate the data collected by Yoshikawa in view of Lehmann in further view of Makoto, Loriaux and Spears using the data validation taught by Anthony because validating this data helps train the machine learning model more accurately (Anthony: “Subsets of the sensors filters, and neural networks diagnose vehicle faults or other problems by building unique machine learning models for each vehicle. Because the approach can adapt, it can provide improvements in the quality of diagnostics over time” [0006]).
Claims 5 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Yoshikawa in view of Lehmann in further view of Makoto, Spears, Loriaux and further in view of Thakur et al. (US PGPUB No US2017/0344941 A1) (hereafter referred to as Thakur).
Regarding claim 5, Yoshikawa in view of Lehmann and in further view of Makoto, Loriaux and Spears discloses all of the limitations of claim 18 as shown in the rejection above. Yoshikawa in view of Lehmann and in further view of Makoto, Spears, Loriaux fails to disclose:
wherein the predicted trip start time of the driving trip of the particular individual is identified by: identifying a match between the initial data and at least a portion of the historical data;
identifying a historical driving trip corresponding to the portion of the historical driving data; and
identifying a start time of the historical driving trip, wherein the start time of the historical driving trip corresponds to the predicted trip start time
However, Thakur teaches:
wherein the predicted trip start time of the driving trip of the particular individual is identified by: identifying a match between the initial data and at least a portion of the historical data (Thakur: “The first location and the first time is received from a service user” [0057]… “The method 4000 associated the first location with a first cell identifier (i.e., initial data) of a grid at 4200. The grid comprises cells, and at least some of the cells are associated with trips previously taken by a vehicle.” [0058] … “upon a condition that the first cell identifier is equal to one of the start cell identifier or the end cell identifier for any single travel segment (i.e., matches at least a portion of the historical data), the method 4000 advances to 4400 to determine whether the first time overlaps the start time associated with one of the start cell identifier or the end cell identifier” [0065]);
identifying a historical driving trip corresponding to the portion of the historical data; and identifying a start time of the historical driving trip (Thakur: “FIG. 5C is a table including a number of travel segments of a vehicle generated from a driving history of the vehicle … The final two columns respectively list the start time range and the end time range for the travel segments (i.e., a start time of the historical trip was identified).” [0064][FIG. 5C]),
wherein the start time of the historical driving trip corresponds to the predicted start time (Thakur: “the method 4000 advances to 4400 to determine whether the first time overlaps the start time or the end time associated with one of the start cell identifier or the end cell identifier (i.e., a predicted range for starting the trip). For example, the first time overlaps the start time or the end time, if they are each single times, if the first time is equal to the start time … within a few minutes” [0066] … “the vehicle is identified in a list of candidate vehicles for the service (i.e., driving trip) associated with the first location and the first time (i.e., the identified start cell identifier and end cell identifier become the predicted interval for the trip, therefore, the start time of the historical trip becomes (i.e., corresponds to) the predicted start time)” [0067]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Lehmann, Spears in further view of Makoto, Loriaux and Thakur. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data, Lehmann discloses training a machine learning model on driver data, and Makoto discloses collecting data over an interval from a start time to an end time. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. Thakur discloses identifying a match between the current trip and a historical trip in order to predict the start time. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. Yoshikawa in view of Lehmann, Spears in further view of Makoto, Loriaux and Thakur both disclose a method of predicting a start time of a driving trip. A person having ordinary skill in the art before the effective filing date of the claimed invention would have recognized that the statistical processing disclosed by Yoshikawa in view of Lehmann, Spears and in further view of Makoto, Loriaux could have been substituted for the process of identifying a match between initial and historical driving trips disclosed by Thakur because both the statistical processing and the match identification process serve the purpose of determining the predicted trip start time. Furthermore, a person having ordinary skill in the art would have been capable of carrying out this substitution. Finally, the substitution achieves the predictable result of determining the predicted trip start time. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to substitute the statistical processing disclosed by Yoshikawa in view of Lehmann and in further view of Makoto, Loriaux for the match identification process disclosed by Thakur according to known methods to yield the predictable result of determining predicted trip start time.
Regarding claim 14, Yoshikawa in view of Lehmann, Spears and in further view of Makoto and Loriaux discloses all of the limitations of claim 18 as shown in the rejection above. Yoshikawa in view of Lehmann, Spears and in further view of Makoto and Loriaux fails to disclose:
wherein the predicted trip start time of the driving trip of the particular individual is identified by: identifying a match between the initial data and at least a portion of the historical data;
identifying a historical driving trip corresponding to the portion of the historical driving data; and
identifying a start time of the historical driving trip, wherein the start time of the historical driving trip corresponds to the predicted trip start time
However, Thakur teaches:
wherein the predicted trip start time of the driving trip of the particular individual is identified by: identifying a match between the initial data and at least a portion of the historical data (Thakur: “The first location and the first time is received from a service user” [0057]… “The method 4000 associated the first location with a first cell identifier (i.e., initial data) of a grid at 4200. The grid comprises cells, and at least some of the cells are associated with trips previously taken by a vehicle.” [0058] … “upon a condition that the first cell identifier is equal to one of the start cell identifier or the end cell identifier for any single travel segment (i.e., matches at least a portion of the historical data), the method 4000 advances to 4400 to determine whether the first time overlaps the start time associated with one of the start cell identifier or the end cell identifier” [0065]);
identifying a historical driving trip corresponding to the portion of the historical data; and identifying a start time of the historical driving trip (Thakur: “FIG. 5C is a table including a number of travel segments of a vehicle generated from a driving history of the vehicle … The final two columns respectively list the start time range and the end time range for the travel segments (i.e., a start time of the historical trip was identified).” [0064][FIG. 5C]),
wherein the start time of the historical driving trip corresponds to the predicted start time (Thakur: “the method 4000 advances to 4400 to determine whether the first time overlaps the start time or the end time associated with one of the start cell identifier or the end cell identifier (i.e., a predicted range for starting the trip). For example, the first time overlaps the start time or the end time, if they are each single times, if the first time is equal to the start time … within a few minutes” [0066] … “the vehicle is identified in a list of candidate vehicles for the service (i.e., driving trip) associated with the first location and the first time (i.e., the identified start cell identifier and end cell identifier become the predicted interval for the trip, therefore, the start time of the historical trip becomes (i.e., corresponds to) the predicted start time)” [0067]).
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Yoshikawa in view of Lehmann, Spears in further view of Makoto, Loriaux and Thakur. Yoshikawa discloses using statistical processing to predict the start time of a trip in order to direct a device to begin collecting driver data, Lehmann discloses training a machine learning model on driver data, and Makoto discloses collecting data over an interval from a start time to an end time. Spears teaches systems and methods to track locations of a mobile device as a function of time and extract routes traversed by the mobile device, frequencies of the routes, and travel time periods of the routes. Thakur discloses identifying a match between the current trip and a historical trip in order to predict the start time. Loriaux teaches the provision of point-of-interest information and, more particularly, to providing users with notifications of convenient purchase points along likely driving routes. Yoshikawa in view of Lehmann in further view of Makoto, Spears, Loriaux and Thakur both disclose a method of predicting a start time of a driving trip. A person having ordinary skill in the art before the effective filing date of the claimed invention would have recognized that the statistical processing disclosed by Yoshikawa in view of Lehmann, Spears and in further view of Makoto, Loriaux could have been substituted for the process of identifying a match between initial and historical driving trips disclosed by Thakur because both the statistical processing and the match identification process serve the purpose of determining the predicted trip start time. Furthermore, a person having ordinary skill in the art would have been capable of carrying out this substitution. Finally, the substitution achieves the predictable result of determining the predicted trip start time. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to substitute the statistical processing disclosed by Yoshikawa in view of Lehmann and in further view of Makoto and Loriaux for the match identification process disclosed by Thakur according to known methods to yield the predictable result of determining predicted trip start time.
Prior Art of Record
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
CN 104715630 (hereafter referred to as Chen) discloses wherein inputting the driving trip data into the machine learning model causes output of a predicted trip end time of a driving trip causes output of a predicted trip end time of a driving trip of the particular individual (see [0016])
CA 3077984 A1 (hereafter referred to as Wang) discloses methods and systems for training and utilizing machine learning models for the purpose of estimating time of arrival (see [0091]).
JP 2011058802 A (hereafter referred to as Miura) (the citations for Miura refer to the machine translated foreign patent document furnished with this official correspondence) discloses methods and systems for the purpose of estimating a route given starting information and utilizing past travel history (see [0045]).
Conclusion
THIS ACTION IS MADE FINAL. 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 AMY TRAN whose telephone number is (571)270-0693. The examiner can normally be reached Monday - Friday 7:30 am - 5:00 pm 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, David Yi can be reached at (571) 272-7519. 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.
/AMY TRAN/Examiner, Art Unit 2126
/DAVID YI/Supervisory Patent Examiner, Art Unit 2126
Prosecution Insights