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 .
Claims 1-20 are pending for examination.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1-20 are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Kobilarov (US 11891088 B1).
Regarding claim 1, Kobilarov teaches a method for generating training data for a machine learning model comprising:
accessing a plurality of output values of a machine learning model computed in response to a plurality of input data samples (Fig. 5: input scenario data to machine learning model and generating an output);
analyzing the plurality of output values and the plurality of input data samples to compute a plurality of required data sample characteristics associated with at least one unsatisfactory output value of the plurality of output values (a performance component 234 may determine a reward 246 (e.g., score) associated with how the vehicle 202 performed in the simulation (and/or in real-world operation, if the planning component's instructions are being used to control the vehicle 202 in real-world operation). In some examples, the reward 246 may be based at least in part on how close the vehicle 202 is to violating an operating constraint and/or how close the vehicle 202 is to an operating target (e.g., maintaining a heading and/or velocity, distance from a target path or position). [col 17: lines 10-20]);
generating at least one new input data sample by providing a data generator with a plurality of generation constraints comprising the plurality of required data sample characteristics (At operation 508, example process 500 may comprise determining a modification to the scenario data based at least in part on a reward determined for performance of the vehicle [Col 23: lines 55-60],); and
adding the at least one new input data sample to a data repository for producing training data for the machine learning model (Operation 610 may comprise modifying an ML component of the vehicle based at least in part on the first reward determined at operation 518. For example, operation 610 may be part of a reinforcement learning algorithm or may be used as part of another ML training technique. [Col 31: lines 42-48]);
wherein the at least one new input data sample comprises at least part of a simulated driving environment for training the machine learning model to operate in an autonomous automotive system (FIG. 5 illustrates a flow diagram of an example process for generating an adversarial agent component, modifying scenario data, and simulating motion of a dynamic object in a simulated environment using the adversarial agent component and/or modified scenario data.).
Regarding claim 2, Kobilarov teaches the method of claim 1, wherein the autonomous automotive system is one or more of: an autonomous driving system (ADS), and an advanced driver-assistance system (ADAS) (Fig.1 : autonomous vehicle).
Regarding claim 3, Kobilarov teaches the method of claim 1, wherein generating the at least one new input data sample comprises modifying at least one of the plurality of input data samples (Fig. 5: determining a modification to the scenario data based at least in part on a reward determined for performance of the vehicle).
Regarding claim 4, Kobilarov teaches the method of claim 1, further comprising:
computing a plurality of performance scores using the plurality of output values and the plurality of input data samples (the techniques may include scoring the autonomous vehicle's performance as it encounters various behaviors and/or scenarios. [Col 2: lines 19-30]);
wherein computing the plurality of required data sample characteristics is further according to the plurality of performance scores (modifying scenario data to reduce the score/reward. Scenario data may specify a configuration, conditions, and/or parameters used by a simulation component to set up and/or run a simulation. [Col 2: lines 44-50]).
Regarding claim 5, Kobilarov teaches the method of claim 4, wherein the plurality of performance scores comprises at least one of: an accuracy score, a precision score, a recall score, an F1 score, and an area under the receiver operating characteristic (ROC) curve (determining the modification based at least in part on using a neural network or other ML model to determine the modification and a confidence score associated with the modification, where the confidence score indicates a likelihood (e.g., a posterior probability) that the modification will decrease the reward. [Col 25: lines 1-10]).
Regarding claim 6, Kobilarov teaches the method of claim 1, wherein the data repository stores a plurality of input data candidates; and
wherein the method further comprises producing the plurality of input data samples by selecting from the data repository a plurality of input data candidates according to a plurality of curation constraints (operation 508 may be used to mine hard examples from a library of scenario data [Col 25: lines 26-28]).
Regarding claim 7, Kobilarov teaches the method of claim 6, wherein the plurality of curation constraints includes at least one target statistical distribution in a set of input data samples of a set of parameter values of a coverage parameter (FIG. 2 illustrates a block diagram of an example architecture for generating an adversarial agent component for use in a simulation component, modifying scenario data to increase the coverage of machine-learned component(s), and/or identifying prospective actions and/or scenarios that may increase a likelihood of vehicle violation of an operating constraint. [Col 1: lines 30-36]).
Regarding claim 8, Kobilarov teaches the method of claim 7, wherein the coverage parameter is one of a set of coverage parameters consisting of:
a class of an object, an object attribute, an object attribute value, a data source, a temporal attribute value, a location attribute value, a difficulty classification of an input data sample, an augmentation technique used to create an input data sample, a sharpness value, a contrast value, a color value, a color combination, a color intensity value, a color brightness, a texture, a histogram of a digital image, a distance between objects, an object orientation, an object orientation relative to another object, an amount of objects in an input data sample, a weather attribute value, a velocity of an object, and a motion pattern of an object (the prospective scenario may be a scenario (e.g., configuration or condition of a roadway; weather condition; positioning, speed, type, action, etc. of an object) or an action of the adversarial agent that is associated with a simulation where the simulated representation of the autonomous vehicle violated an operating constraint [Col: 52-60]).
Regarding claim 9, Kobilarov teaches the method of claim 6, wherein the plurality of curation constraints includes at least one qualitative characteristic of a set of input data samples (the prospective scenario may be a scenario (e.g., configuration or condition of a roadway; weather condition; positioning, speed, type, action, etc. of an object) or an action of the adversarial agent that is associated with a simulation where the simulated representation of the autonomous vehicle violated an operating constraint [Col: 52-60] Examiner note: non limiting examples of qualitative data).
Regarding claim 10, Kobilarov teaches the method of claim 9, wherein the at least one qualitative characteristic comprises at least one of: a semantic context of an object, a composition of a plurality of objects, a plurality of parameter values of a plurality of coverage parameters in an input data sample, a variation between a plurality of compositions of a plurality of objects in the set of input data samples, and a rarity value of a composition of a plurality of objects (the prospective scenario may be a scenario (e.g., configuration or condition of a roadway; weather condition; positioning, speed, type, action, etc. of an object) or an action of the adversarial agent that is associated with a simulation where the simulated representation of the autonomous vehicle violated an operating constraint [Col: 52-60]).
Regarding claim 11, Kobilarov teaches the method of claim 6, further comprising modifying the plurality of curation constraints according to the plurality of required data sample characteristics (modification(s) to the scenario data to determine a prospective scenario/event that would be difficult for the autonomous vehicle to handle. For example, the prospective scenario may be a scenario (e.g., configuration or condition of a roadway; weather condition; positioning, speed, type, action, etc. of an object) or an action of the adversarial agent that is associated with a simulation where the simulated representation of the autonomous vehicle violated an operating constraint. [Col: 52-60]).).
Regarding claim 12, Kobilarov teaches the method of claim 1, further comprising:
collecting raw data from a plurality of data sources; and adding to the data repository at least one other input data sample generated using at least some of the raw data (the example scenario 100 may be based at least in part on log data received from one or more autonomous vehicles (i.e., data captured and stored during actual driving in a physical environment including raw sensor data and data derived therefrom including detections, predictions, control signals, etc.). [col 5: lines 17-25]).
Regarding claim 13, Kobilarov teaches the method of claim 12, wherein the plurality of data sources comprises at least one of: a database, a sensor, an application programming interface and a human-machine interface (the example scenario 100 may be based at least in part on log data received from one or more autonomous vehicles (i.e., data captured and stored during actual driving in a physical environment including raw sensor data and data derived therefrom including detections, predictions, control signals, etc.). [col 5: lines 17-25]).
Regarding claim 14, Kobilarov teaches the method of claim 1, further comprising training the machine learning model using one or more sets of training data selected from the data repository (Operation 610 may comprise modifying an ML component of the vehicle based at least in part on the first reward determined at operation 518. For example, operation 610 may be part of a reinforcement learning algorithm or may be used as part of another ML training technique. [Col 31: lines 42-48])
Regarding claim 15, Kobilarov teaches the method of claim 14, wherein the data repository stores a plurality of input data candidates; and
wherein at least one of the one or more sets of training data is produced by selecting from the data repository another plurality of input data candidates according to another plurality of curation constraints (operation 508 may be used to mine hard examples from a library of scenario data and/or operation 508 may be used to iteratively make subsequent simulations more and more difficult until the simulated vehicle violates an operating constraint. Regardless, the modified scenario data may be stored and/or used to instantiate a simulation. [Col 25: lines 25-32]).
Regarding claim 16, Kobilarov teaches the method of claim 1, wherein the at least one new input data sample comprises at least one of: a digital image, a digital video, and a simulated signal simulating a signal captured from a sensor (The scenario data structure may comprise a two-dimensional image, a publish-subscribe message, and/or the like. [Col: 24 Lines: 18-20]).
Regarding claim 17, Kobilarov teaches the method of claim 1, further comprising:
computing a plurality of visual features by analyzing an identified repository of data samples comprising a plurality of digital images captured in one or more physical environments (the perception component 112 may receive sensor data associated with the environment and determine a position, orientation, velocity, classification (e.g., vehicle, sedan, “passenger vehicle”), track (e.g., historical, current, and/or predicted attributes, such as position, orientation, velocity, etc.), etc. of the object 120 and/or object 122. Perception data generated by the perception component 112 and/or sensor data may be stored in a log data store, which may comprise the position and/or other attributes of object 120 and/or object 122 at a current, previous, and/or future time steps, where the future time steps may be associated with predicted data. [Col 7: lines 45-55]); and
computing at least one dataset score using the plurality of visual features and at least one additional set of training data selected from the data repository, where the at least one dataset score is indicative of an expected performance score of the machine learning model in response to input data when the machine learning model is trained using the at least one additional set of training data (The vehicle 102 may transmit at least part of the sensor data, perception data, and/or a scenario label (if one is generated on-vehicle by the perception component 112) as part of the log data to a remote computing device (unillustrated in FIG. 1).);
wherein computing the plurality of required data sample characteristics is further according to the at least one dataset score and the plurality of visual features (Fig. 5).
Regarding claim 18, Kobilarov teaches the method of claim 17, wherein the plurality of visual features comprises at least one of: an object attribute, an object attribute value, a sharpness value, a contrast value, a color value, a color combination, a color intensity value, a color brightness, a texture, a histogram of a digital image, a distance between objects, an object orientation, an object orientation relative to another object, an amount of objects in an input data sample, a weather attribute value, a velocity of an object, a semantic context of an object, an edge of an object, a positional relationship between two or more objects, and a motion pattern of an object (modification(s) to the scenario data to determine a prospective scenario/event that would be difficult for the autonomous vehicle to handle. For example, the prospective scenario may be a scenario (e.g., configuration or condition of a roadway; weather condition; positioning, speed, type, action, etc. of an object) or an action of the adversarial agent that is associated with a simulation where the simulated representation of the autonomous vehicle violated an operating constraint. [Col: 52-60]).).
Regarding claim 19, the system of claim 19 is rejected under the same arts and evidence used to reject claims 1-18.
Regarding claim 20, the software product of claim 20 is rejected under the same arts and evidence used to reject claims 1-18.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HESHAM K ABOUZAHRA whose telephone number is (571)270-0425. The examiner can normally be reached M-F 8-5.
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, Jamie Atala can be reached at 57127227384. 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.
/HESHAM K ABOUZAHRA/ Primary Examiner, Art Unit 2486
Prosecution Insights