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 .
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.
Information Disclosure Statement
The information disclosure statement(s) (IDS) submitted on 10/25/2025 is/are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement(s) is/are being considered by the examiner.
Priority
The applicant's claim for benefit of Provisional Patent Application Serial No. 63/527,486 filed on 07/18/2023 has been received and acknowledged.
Claim Objections
Claims 4, 13, and 20 are objected to because of the following informalities:
"wherein the ad hoc control circuit configured to" should read "wherein the ad hoc control circuit is configured to" (claim 4)
"the one or other vehicles" should read "the one or more other vehicles" (Claim 4)
"wherein determining the composite identification comprises as a function of" should read "wherein determining the composite identification is a function of" (claims 13 and 20)
Appropriate correction is required.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1, 2, 3, 4, 10, 11, 12, 13, 19, 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over US 20190212746 A1 Cheng; Bin et al. (hereinafter Cheng), in view of US 20200210721 A1 Goel; Kratarth et al. (hereinafter Goel).
Regarding claim 1, Cheng discloses: A system (see Cheng at least [0002] a sensor system for multiple perspective sensor data sets), comprising:
a vehicle control circuit (see Cheng at least [0084] In some embodiments, the sensor system 199 of the vehicle 123 may be implemented using hardware including a field-programmable gate array (“FPGA”) or an application-specific integrated circuit (“ASIC”)) configured to determine, using a machine learning model executed by the vehicle control circuit and information from one or more sensors of the vehicle (see Cheng at least [0105] the sensor module 204 includes a deep learning algorithm that analyzes images included in the sensor data to construct preliminary heatmaps and bounding boxes around objects included in the preliminary heatmaps):
one or more intermediate classifications and respective confidence indications for the one or more determined intermediate classifications (see Cheng at least [0110] the preliminary voxel maps and preliminary heatmaps are electronic maps that describe a preliminary estimate of a position of the objects within a three-dimensional (3D) environment of the particular vehicle 123, 124 (e.g., the roadway environment), a preliminary classification for each object (e.g., car, road, bicycle, pedestrian, etc.) and a preliminary confidence value that the preliminary classification is correct); and
an ad hoc control circuit configured to receive one or more intermediate classifications from a dynamic ad hoc network, the dynamic ad hoc network comprising one or more other vehicles (see Cheng at least [0006] the sensor system generates digital data that describes one or more of the following: sensor measurements; objects that are indicated by the sensor measurements; type classifications for these detected objects (e.g., car, road, bicycle, pedestrian, etc.); and confidence values in the type classifications. In some embodiments, the sensor system wirelessly shares the digital data with other automated vehicles via a wireless network. For example, the sensor system is operable to wirelessly share the digital data with other automated vehicles via V2V, V2X, or other intervehicle communication technologies (e.g. 5G, LTE, etc.)), wherein the vehicle control circuit is configured to:
determine a composite identification as a function of the one or more received intermediate classifications from the dynamic ad hoc network and at least one of the one or more determined intermediate classifications by the vehicle control circuit or the determined first identification by the vehicle control circuit (see Cheng at least [0013] reconcile discrepancies in the set of heatmaps and a preliminary heatmap of the ego vehicle to form a combined heatmap that describes the objects in the roadway environment as collectively observed by the onboard sensors of the set of remote vehicles and the ego vehicle and [0121] the localization module 208 modifies the combined heatmap data so that the combined heatmap data describes, for each significant object, (1) the exact position of the significant object in 3D space and (2) the most likely classification for each significant object); and
determine one or more control inputs to the vehicle based at least in part on the determined composite identification (see Cheng at least [0011] modify an operation of the ego vehicle based on the combined heatmap).
Cheng does not teach: a first identification with a first confidence indication using at least one of the one or more determined intermediate classifications.
However, Goel teaches: a first identification with a first confidence indication using at least one of the one or more determined intermediate classifications (see Goel at least [0011] a selection component may provide a first subset to a first sub-class ML model based at least in part on a first classification and [0042] In some examples, an individual sub-class ML model 304(p) may be trained to output (310) a sub-classification and/or a probability from among one or more candidate sub-classifications that are associated with the general classification with which the sub-class ML model 304(p) is associated).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the inter-vehicle communication and classification system disclosed by Cheng to include the two-stage ego vehicle object identification process of Goel. One of ordinary skill in the art would have been motivated to make this modification because refining an object classification intermediate result through further analysis can increase the accuracy and/or confidence levels associated with the object detections, as suggested by Goel (see Goel at least [0019] The techniques described herein may improve the accuracy of detections of objects by increasing the specificity with which an object may be classified and/or increasing a confidence score generated by the first ML model and/or the sub-class ML model in association with an object detection).
Regarding claim 2, Cheng and Goel disclose: The system of claim 1, wherein the one or more determined intermediate classifications include one or more determined intermediate classifications of an object and respective confidence indications for the one or more determined intermediate classifications of the object (see Cheng at least [0009] the heatmaps included in the set of preliminary heatmaps describes objects in the roadway environment),
wherein the first identification includes a first identification of the object with the first confidence indication using the one or more determined intermediate classifications of the object (see Cheng at least [0110] the preliminary voxel maps and preliminary heatmaps are electronic maps that describe a preliminary estimate of a position of the objects within a three-dimensional (3D) environment of the particular vehicle 123, 124 (e.g., the roadway environment), a preliminary classification for each object (e.g., car, road, bicycle, pedestrian, etc.) and a preliminary confidence value that the preliminary classification is correct),
wherein the one or more received intermediate classifications from one or more other vehicles in the dynamic ad hoc network include received intermediate classifications of the object from the one or more other vehicles in the dynamic ad hoc network (see Cheng at least [0006] the sensor system generates digital data that describes one or more of the following: sensor measurements; objects that are indicated by the sensor measurements; type classifications for these detected objects (e.g., car, road, bicycle, pedestrian, etc.); and confidence values in the type classifications. In some embodiments, the sensor system wirelessly shares the digital data with other automated vehicles via a wireless network. For example, the sensor system is operable to wirelessly share the digital data with other automated vehicles via V2V, V2X, or other intervehicle communication technologies (e.g. 5G, LTE, etc.)).
Regarding claim 3, Cheng and Goel disclose: The system of claim 2, comprising the one or more sensors of the vehicle including at least one of a camera, an ultrasonic transducer, a LiDAR sensor, or a radar sensor (see Cheng at least [0071] the sensor set 195 of the vehicle 123 may include one or more of the following vehicle sensors: a vibrometer; a collision detection system; an engine oil pressure detection sensor; a camera (e.g., one or more of an internal camera and an external camera); a LIDAR sensor; an ultrasonic sensor; a radar sensor),
wherein the vehicle control circuit is configured to receive the information from the one or more sensors (see Cheng at least [0107] each of the ego vehicle and the remote vehicles collects their own sensor measurements), detect the object in the received information (see Cheng at least [0006] the sensor system generates digital data that describes one or more of the following: … objects that are indicated by the sensor measurements), and to determine the intermediate classifications using information about the detected object from the one or more sensors (see Cheng at least [0006] the sensor system generates digital data that describes one or more of the following: … type classifications for these detected objects (e.g., car, road, bicycle, pedestrian, etc.)),
wherein the object includes an object within a sensor range of the one or more sensors of the vehicle (see Cheng at least [0047] vehicles that are within sensor range of these distance vehicle), including one or more of a line or an indication of a lane of travel of a roadway of the vehicle, one or more other vehicles (see Cheng at least [0155] each object detected (e.g. vehicle or pedestrian)), an item on or proximate to the roadway, or a traffic sign above or proximate to the roadway.
Regarding claim 4, Cheng and Goel disclose: The system of claim 1, wherein the ad hoc control circuit configured to receive intermediate classifications and respective confidence indications for the one or more received intermediate classifications from the one or more other vehicles in the dynamic ad hoc network (see Cheng at least [0145] the van is not detected by the automated driving system of V1 and detected with low confidence by the automated driving system of vehicles V2 and V3. After the fusion module 206 is executed and the results of this execution are shared via wireless communication with the other vehicles),
wherein the vehicle control circuit is configured to determine the composite identification as a function of at least one of the one or more determined intermediate classifications and respective confidence indications by the vehicle control circuit and at least one of the one or more received intermediate classifications and respective confidence indications from the one or other vehicles in the dynamic ad hoc network (see Cheng at least [0116] generate a combined heatmap that visually depicts each object and the confidence that the class assigned to each object in step (4) is correct).
Regarding claim 10, Cheng discloses: A method (see Cheng at least [0009] a method), comprising:
determining, using a machine learning model executed by a vehicle control circuit (see Cheng at least [0084] In some embodiments, the sensor system 199 of the vehicle 123 may be implemented using hardware including a field-programmable gate array (“FPGA”) or an application-specific integrated circuit (“ASIC”)) and information from one or more sensors of the vehicle (see Cheng at least [0105] the sensor module 204 includes a deep learning algorithm that analyzes images included in the sensor data to construct preliminary heatmaps and bounding boxes around objects included in the preliminary heatmaps), one or more intermediate classifications and respective confidence indications for the one or more determined intermediate classifications (see Cheng at least [0110] the preliminary voxel maps and preliminary heatmaps are electronic maps that describe a preliminary estimate of a position of the objects within a three-dimensional (3D) environment of the particular vehicle 123, 124 (e.g., the roadway environment), a preliminary classification for each object (e.g., car, road, bicycle, pedestrian, etc.) and a preliminary confidence value that the preliminary classification is correct);
receiving one or more intermediate classifications from a dynamic ad hoc network, the dynamic ad hoc network comprising one or more other vehicles (see Cheng at least [0006] the sensor system generates digital data that describes one or more of the following: sensor measurements; objects that are indicated by the sensor measurements; type classifications for these detected objects (e.g., car, road, bicycle, pedestrian, etc.); and confidence values in the type classifications. In some embodiments, the sensor system wirelessly shares the digital data with other automated vehicles via a wireless network. For example, the sensor system is operable to wirelessly share the digital data with other automated vehicles via V2V, V2X, or other intervehicle communication technologies (e.g. 5G, LTE, etc.));
determining a composite identification as a function of the one or more received intermediate classifications from the dynamic ad hoc network and at least one of the one or more determined intermediate classifications by the vehicle control circuit or the determined first identification by the vehicle control circuit (see Cheng at least [0013] reconcile discrepancies in the set of heatmaps and a preliminary heatmap of the ego vehicle to form a combined heatmap that describes the objects in the roadway environment as collectively observed by the onboard sensors of the set of remote vehicles and the ego vehicle and [0121] the localization module 208 modifies the combined heatmap data so that the combined heatmap data describes, for each significant object, (1) the exact position of the significant object in 3D space and (2) the most likely classification for each significant object); and
determining one or more control inputs to the vehicle based at least in part on the determined composite identification (see Cheng at least [0011] modify an operation of the ego vehicle based on the combined heatmap).
Cheng does not teach: determining a first identification with a first confidence indication using at least one of the one or more determined intermediate classifications.
However, Goel teaches: determining a first identification with a first confidence indication using at least one of the one or more determined intermediate classifications (see Goel at least [0011] a selection component may provide a first subset to a first sub-class ML model based at least in part on a first classification and [0042] In some examples, an individual sub-class ML model 304(p) may be trained to output (310) a sub-classification and/or a probability from among one or more candidate sub-classifications that are associated with the general classification with which the sub-class ML model 304(p) is associated).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the inter-vehicle communication and classification method disclosed by Cheng to include the two-stage ego vehicle object identification process of Goel. One of ordinary skill in the art would have been motivated to make this modification because refining an object classification intermediate result through further analysis can increase the accuracy and/or confidence levels associated with the object detections, as suggested by Goel (see Goel at least [0019] The techniques described herein may improve the accuracy of detections of objects by increasing the specificity with which an object may be classified and/or increasing a confidence score generated by the first ML model and/or the sub-class ML model in association with an object detection).
Regarding claim 11, Cheng and Goel disclose: The method of claim 10, wherein determining the one or more intermediate classifications include determining one or more intermediate classifications of an object and respective confidence indications for the one or more determined intermediate classifications of the object (see Cheng at least [0009] the heatmaps included in the set of preliminary heatmaps describes objects in the roadway environment),
wherein determining the first identification includes determining a first identification of the object with the first confidence indication using at least one of the one or more determined intermediate classifications of the object (see Cheng at least [0110] the preliminary voxel maps and preliminary heatmaps are electronic maps that describe a preliminary estimate of a position of the objects within a three-dimensional (3D) environment of the particular vehicle 123, 124 (e.g., the roadway environment), a preliminary classification for each object (e.g., car, road, bicycle, pedestrian, etc.) and a preliminary confidence value that the preliminary classification is correct),
wherein receiving one or more intermediate classifications from the dynamic ad hoc network includes receiving one or more intermediate classifications of the object from the one or more other vehicles in the dynamic ad hoc network (see Cheng at least [0006] the sensor system generates digital data that describes one or more of the following: sensor measurements; objects that are indicated by the sensor measurements; type classifications for these detected objects (e.g., car, road, bicycle, pedestrian, etc.); and confidence values in the type classifications. In some embodiments, the sensor system wirelessly shares the digital data with other automated vehicles via a wireless network. For example, the sensor system is operable to wirelessly share the digital data with other automated vehicles via V2V, V2X, or other intervehicle communication technologies (e.g. 5G, LTE, etc.)).
Regarding claim 12, Cheng and Goel disclose: The method of claim 11, comprising:
receiving information from the one or more sensors of the vehicle (see Cheng at least [0107] each of the ego vehicle and the remote vehicles collects their own sensor measurements), the one or more sensors of the vehicle including at least one of a camera, an ultrasonic transducer, a LiDAR sensor, or a radar sensor (see Cheng at least [0071] the sensor set 195 of the vehicle 123 may include one or more of the following vehicle sensors: a vibrometer; a collision detection system; an engine oil pressure detection sensor; a camera (e.g., one or more of an internal camera and an external camera); a LIDAR sensor; an ultrasonic sensor; a radar sensor),
wherein determining the first identification of the object includes detecting the object in the received information (see Cheng at least [0006] the sensor system generates digital data that describes one or more of the following: … objects that are indicated by the sensor measurements),
wherein the object includes an object within a sensor range of the one or more sensors of the vehicle (see Cheng at least [0047] vehicles that are within sensor range of these distance vehicle), including one or more of a line or an indication of a lane of travel of a roadway of the vehicle, one or more other vehicles (see Cheng at least [0155] each object detected (e.g. vehicle or pedestrian)), an item on or proximate to the roadway, or a traffic sign above or proximate to the roadway.
Regarding claim 13, Cheng and Goel disclose: The method of claim 10, wherein receiving one or more intermediate classifications and respective confidence indications for the one or more received intermediate classifications from the dynamic ad hoc network comprises receiving one or more intermediate classifications and respective confidence indications for the one or more received intermediate classifications from the one or more other vehicles in the dynamic ad hoc network (see Cheng at least [0145] the van is not detected by the automated driving system of V1 and detected with low confidence by the automated driving system of vehicles V2 and V3. After the fusion module 206 is executed and the results of this execution are shared via wireless communication with the other vehicles),
wherein determining the composite identification comprises as a function of at least one of the determined intermediate classifications and respective confidence indications by the vehicle control circuit and at least one of the one or more received intermediate classifications and respective confidence indications from the one or other vehicles in the dynamic ad hoc network (see Cheng at least [0116] generate a combined heatmap that visually depicts each object and the confidence that the class assigned to each object in step (4) is correct).
Regarding claim 19, Cheng discloses: A system (see Cheng at least [0002] a sensor system for multiple perspective sensor data sets), comprising:
one or more processors (see Cheng at least [0077] the processor 125); and
a memory storing computer-executable instructions that, when executed, cause the one or more processors to control the system to perform operations (see Cheng at least [0077] The memory 127 stores instructions or data that may be accessed and executed by the processor 125. The instructions or data may include code for performing the techniques described herein) comprising:
determining, using a machine learning model executed by a vehicle control circuit (see Cheng at least [0084] In some embodiments, the sensor system 199 of the vehicle 123 may be implemented using hardware including a field-programmable gate array (“FPGA”) or an application-specific integrated circuit (“ASIC”)) and information from one or more sensors of the vehicle (see Cheng at least [0105] the sensor module 204 includes a deep learning algorithm that analyzes images included in the sensor data to construct preliminary heatmaps and bounding boxes around objects included in the preliminary heatmaps) intermediate classifications and respective confidence indications of the determined intermediate classifications (see Cheng at least [0110] the preliminary voxel maps and preliminary heatmaps are electronic maps that describe a preliminary estimate of a position of the objects within a three-dimensional (3D) environment of the particular vehicle 123, 124 (e.g., the roadway environment), a preliminary classification for each object (e.g., car, road, bicycle, pedestrian, etc.) and a preliminary confidence value that the preliminary classification is correct);
receiving one or more intermediate classifications from a dynamic ad hoc network, the dynamic ad hoc network comprising one or more other vehicles (see Cheng at least [0006] the sensor system generates digital data that describes one or more of the following: sensor measurements; objects that are indicated by the sensor measurements; type classifications for these detected objects (e.g., car, road, bicycle, pedestrian, etc.); and confidence values in the type classifications. In some embodiments, the sensor system wirelessly shares the digital data with other automated vehicles via a wireless network. For example, the sensor system is operable to wirelessly share the digital data with other automated vehicles via V2V, V2X, or other intervehicle communication technologies (e.g. 5G, LTE, etc.));
determining a composite identification as a function of at least one of the one or more received intermediate classifications from the dynamic ad hoc network and at least one of the one or more determined intermediate classifications by the vehicle control circuit or the determined first identification by the vehicle control circuit (see Cheng at least [0013] reconcile discrepancies in the set of heatmaps and a preliminary heatmap of the ego vehicle to form a combined heatmap that describes the objects in the roadway environment as collectively observed by the onboard sensors of the set of remote vehicles and the ego vehicle and [0121] the localization module 208 modifies the combined heatmap data so that the combined heatmap data describes, for each significant object, (1) the exact position of the significant object in 3D space and (2) the most likely classification for each significant object); and
determining one or more control inputs to the vehicle based at least in part on the determined composite identification (see Cheng at least [0011] modify an operation of the ego vehicle based on the combined heatmap).
Cheng does not teach: determining a first identification with a first confidence indication using the determined intermediate classifications.
However, Goel teaches: determining a first identification with a first confidence indication using the determined intermediate classifications (see Goel at least [0011] a selection component may provide a first subset to a first sub-class ML model based at least in part on a first classification and [0042] In some examples, an individual sub-class ML model 304(p) may be trained to output (310) a sub-classification and/or a probability from among one or more candidate sub-classifications that are associated with the general classification with which the sub-class ML model 304(p) is associated).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the inter-vehicle communication and classification system disclosed by Cheng to include the two-stage ego vehicle object identification process of Goel. One of ordinary skill in the art would have been motivated to make this modification because refining an object classification intermediate result through further analysis can increase the accuracy and/or confidence levels associated with the object detections, as suggested by Goel (see Goel at least [0019] The techniques described herein may improve the accuracy of detections of objects by increasing the specificity with which an object may be classified and/or increasing a confidence score generated by the first ML model and/or the sub-class ML model in association with an object detection).
Regarding claim 20, Cheng and Goel disclose: The system of claim 19, wherein receiving one or more intermediate classifications and respective confidence indications for the one or more received intermediate classifications from the dynamic ad hoc network comprises receiving one or more intermediate classifications and respective confidence indications for the one or more received intermediate classifications from the one or more other vehicles in the dynamic ad hoc network (see Cheng at least [0145] the van is not detected by the automated driving system of V1 and detected with low confidence by the automated driving system of vehicles V2 and V3. After the fusion module 206 is executed and the results of this execution are shared via wireless communication with the other vehicles),
wherein determining the composite identification comprises as a function of at least one of the determined intermediate classifications and respective confidence indications by the vehicle control circuit and at least one of the one or more received intermediate classifications and respective confidence indications from the one or other vehicles in the dynamic ad hoc network (see Cheng at least [0116] generate a combined heatmap that visually depicts each object and the confidence that the class assigned to each object in step (4) is correct).
Claim(s) 5, 7, 14, 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Cheng, in view of Goel, further in view of DE 102021206980 A1 HERRMANN CHRISTIAN et al. (hereinafter Herrmann).
Regarding claim 5, Cheng and Goel disclose: The system of claim 1.
Cheng and Goel do not teach: wherein the ad hoc control circuit is configured to determine a size of the dynamic ad hoc network for the vehicle, the size of the dynamic ad hoc network including a number of other vehicles from which to receive or process intermediate classifications or a range for seeking potential data sources, based on at least one of the determined respective confidence indications for the one or more determined intermediate classifications from the vehicle control circuit.
However, Herrmann teaches: wherein the ad hoc control circuit is configured to determine a size of the dynamic ad hoc network for the vehicle, the size of the dynamic ad hoc network including a number of other vehicles from which to receive or process intermediate classifications or a range for seeking potential data sources, based on at least one of the determined respective confidence indications for the one or more determined intermediate classifications from the vehicle control circuit (see Herrmann at least [pg. 5, para. 2, beginning with “The request can depend”] If, for example, the occurrence of a possible signaling event is subject to a certain degree of uncertainty, a request for further data is sent. One possible scenario provides, for example, that the vehicle cannot clearly assign a specific object to a specific object class, for example to decide whether the object detected in the area surrounding the vehicle is a bag or a stone. The vehicle then sends a request for further classification information, which can be made available by vehicles parked in the area or vehicles driving past and which may only relate to that part of the exterior space in which the object that has already been detected is located).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the inter-vehicle communication and classification system disclosed by Cheng and Goel to include the certainty-dependent requesting of inter-vehicle classification assistance of Herrmann. One of ordinary skill in the art would have been motivated to make this modification because assuming that an ego vehicle is unable to confidently classify an object by using only its own sensors and object recognition capabilities, the ego vehicle is able to request assistance from another vehicle in range to perceive the object only if confidence levels require such an action, as suggested by Herrmann (see Herrmann at least [pg. 5, para. 2, beginning with “The request can depend”] The other vehicles can then transmit corresponding raw image data and/or object classification results, so that the (first) vehicle uses the additional data made available to finally assign the detected object to a specific object class in its control electronics and then, in response to the respective signaling event, assigns a specific function to the vehicle triggers).
Regarding claim 7, Cheng, Goel, and Herrmann disclose: The system of claim 5, comprising a communication circuit configured to connect to the one or more other vehicles in the dynamic ad hoc network (see Cheng at least [0107] In some embodiments, the vehicle 123 and the remote vehicles 124 each include a communication unit 245. The communication unit 245 includes any hardware and software necessary to enable the vehicle 123 and the remote vehicles 124 to transmit wireless messages to one another), wherein the communication circuit is configured to:
transmit at least one of the one or more determined intermediate classifications to the one or more other vehicles in the dynamic ad hoc network (see Cheng at least [0108] cause the communication unit 245 to broadcast or unicast the sensor data to the remote vehicles 124); and
receive the one or more intermediate classifications from the one or more other vehicles in the dynamic ad hoc network (see Cheng at least [0108] The sensor systems 199 of the remote vehicles 124 also share their sensor data with the vehicle 123).
Regarding claim 14, Cheng and Goel disclose: The method of claim 10.
Cheng and Goel do not teach: comprising, determining a size of the dynamic ad hoc network for the vehicle, the size of the dynamic ad hoc network including a number of other vehicles from which to receive or process intermediate classifications or a range for seeking potential data sources, based on at least one of the one or more determined respective confidence indications for the one or more determined intermediate classifications.
However, Herrmann teaches: comprising, determining a size of the dynamic ad hoc network for the vehicle, the size of the dynamic ad hoc network including a number of other vehicles from which to receive or process intermediate classifications or a range for seeking potential data sources, based on at least one of the one or more determined respective confidence indications for the one or more determined intermediate classifications (see Herrmann at least [pg. 5, para. 2, beginning with “The request can depend”] If, for example, the occurrence of a possible signaling event is subject to a certain degree of uncertainty, a request for further data is sent. One possible scenario provides, for example, that the vehicle cannot clearly assign a specific object to a specific object class, for example to decide whether the object detected in the area surrounding the vehicle is a bag or a stone. The vehicle then sends a request for further classification information, which can be made available by vehicles parked in the area or vehicles driving past and which may only relate to that part of the exterior space in which the object that has already been detected is located).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the inter-vehicle communication and classification method disclosed by Cheng and Goel to include the certainty-dependent requesting of inter-vehicle classification assistance of Herrmann. One of ordinary skill in the art would have been motivated to make this modification because assuming that an ego vehicle is unable to confidently classify an object by using only its own sensors and object recognition capabilities, the ego vehicle is able to request assistance from another vehicle in range to perceive the object only if confidence levels require such an action, as suggested by Herrmann (see Herrmann at least [pg. 5, para. 2, beginning with “The request can depend”] The other vehicles can then transmit corresponding raw image data and/or object classification results, so that the (first) vehicle uses the additional data made available to finally assign the detected object to a specific object class in its control electronics and then, in response to the respective signaling event, assigns a specific function to the vehicle triggers).
Regarding claim 16, Cheng, Goel, and Herrmann disclose: The method of claim 14, comprising:
connecting to the one or more other vehicles in the dynamic ad hoc network using a communication circuit (see Cheng at least [0107] In some embodiments, the vehicle 123 and the remote vehicles 124 each include a communication unit 245. The communication unit 245 includes any hardware and software necessary to enable the vehicle 123 and the remote vehicles 124 to transmit wireless messages to one another); and
transmitting at least one of the one or more determined intermediate classifications to one or more other vehicles in the dynamic ad hoc network using the communication circuit (see Cheng at least [0108] cause the communication unit 245 to broadcast or unicast the sensor data to the remote vehicles 124).
Claim(s) 6, 8, 15, 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Cheng, in view of Goel, further in view of Herrmann, and further in view of US 20190139403 A1 Alam; S M Iftekharul et al. (hereinafter Alam).
Regarding claim 6, Cheng, Goel, and Herrmann disclose: The system of claim 5.
Cheng, Goel, and Herrmann do not teach: wherein the ad hoc control circuit is configured to adjust the determined size of the ad hoc network based on a difference between at least one of the one or more determined intermediate classifications by the vehicle control circuit and a corresponding at least one of the one or more received intermediate classifications from the dynamic ad hoc network.
However, Alam teaches: wherein the ad hoc control circuit is configured to adjust the determined size of the ad hoc network based on a difference between at least one of the one or more determined intermediate classifications by the vehicle control circuit and a corresponding at least one of the one or more received intermediate classifications from the dynamic ad hoc network (see Alam at least [0028] one or more differences (e.g., deviations, deltas) between a crowdsourced map of an ambient environment and a real-time volumetric map of the ambient environment and [0030] As already noted, before sharing the delta with others, a size-based classifier that takes the calculated delta as an input and determines whether the delta poses a persistent threat).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the inter-vehicle communication and classification system disclosed by Cheng, Goel, and Herrmann to include the deviation-dependent requesting of inter-vehicle classification assistance of Alam. One of ordinary skill in the art would have been motivated to make this modification because communication with other vehicles should be focused on situations when extra information is necessary, as opposed to times when different sources agree on insignificant detected objects, as suggested by Alam (see Alam at least [0030] Such an approach helps to avoid unnecessary reporting of tiny objects on the road and reduces the burden on communication links between vehicles and the cloud server).
Regarding claim 8, Cheng, Goel, and Herrmann disclose: The system of claim 7.
Cheng, Goel, and Herrmann do not teach: wherein the communication circuit is configured to withhold transmitting the determined first identification to the one or more other vehicles in the dynamic ad hoc network.
However, Alam teaches: wherein the communication circuit is configured to withhold transmitting the determined first identification to the one or more other vehicles in the dynamic ad hoc network (see Alam at least [0030] As already noted, before sharing the delta with others, a size-based classifier that takes the calculated delta as an input and determines whether the delta poses a persistent threat. The classifier deems objects of certain shapes as hazardous because they are large enough to cause problem to driving actions).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the inter-vehicle communication and classification system disclosed by Cheng, Goel, and Herrmann to include the classification-dependent requesting of inter-vehicle classification assistance of Alam. One of ordinary skill in the art would have been motivated to make this modification because communicating with other vehicles based on the importance of the identified object avoids excessive communication and resource-intensive classification of trivial objects, as suggested by Alam (see Alam at least [0030] Such an approach helps to avoid unnecessary reporting of tiny objects on the road and reduces the burden on communication links between vehicles and the cloud server).
Regarding claim 15, Cheng, Goel, and Herrmann disclose: the method of claim 14.
Cheng, Goel, and Herrmann do not teach: comprising: adjusting the determined size of the ad hoc network based on a difference between at least one of the one or more determined intermediate classifications by the vehicle control circuit and a corresponding at least one of the one or more received intermediate classifications from the one or more other vehicles in the dynamic ad hoc network.
However, Alam teaches: comprising: adjusting the determined size of the ad hoc network based on a difference between at least one of the one or more determined intermediate classifications by the vehicle control circuit and a corresponding at least one of the one or more received intermediate classifications from the one or more other vehicles in the dynamic ad hoc network(see Alam at least [0028] one or more differences (e.g., deviations, deltas) between a crowdsourced map of an ambient environment and a real-time volumetric map of the ambient environment and [0030] As already noted, before sharing the delta with others, a size-based classifier that takes the calculated delta as an input and determines whether the delta poses a persistent threat).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the inter-vehicle communication and classification method disclosed by Cheng, Goel, and Herrmann to include the deviation-dependent requesting of inter-vehicle classification assistance of Alam. One of ordinary skill in the art would have been motivated to make this modification because communication with other vehicles should be focused on situations when extra information is necessary, as opposed to times when different sources agree on insignificant detected objects, as suggested by Alam (see Alam at least [0030] Such an approach helps to avoid unnecessary reporting of tiny objects on the road and reduces the burden on communication links between vehicles and the cloud server).
Regarding claim 17, Cheng, Goel, and Herrmann disclose: The method of claim 14.
Cheng, Goel, and Herrmann do not teach: comprising: withholding transmitting the determined first identification to the one or more other vehicles in the dynamic ad hoc network.
However, Alam teaches: comprising: withholding transmitting the determined first identification to the one or more other vehicles in the dynamic ad hoc network (see Alam at least [0030] As already noted, before sharing the delta with others, a size-based classifier that takes the calculated delta as an input and determines whether the delta poses a persistent threat. The classifier deems objects of certain shapes as hazardous because they are large enough to cause problem to driving actions).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the inter-vehicle communication and classification method disclosed by Cheng, Goel, and Herrmann to include the classification-dependent requesting of inter-vehicle classification assistance of Alam. One of ordinary skill in the art would have been motivated to make this modification because communicating with other vehicles based on the importance of the identified object avoids excessive communication and resource-intensive classification of trivial objects, as suggested by Alam (see Alam at least [0030] Such an approach helps to avoid unnecessary reporting of tiny objects on the road and reduces the burden on communication links between vehicles and the cloud server).
Claim(s) 9, 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Cheng, in view of Goel, further in view of Herrmann, and further in view of US 20110006913 A1 Chen; Wai et al. (hereinafter Chen).
Regarding claim 9, Cheng, Goel, and Herrmann disclose: The system of claim 7, wherein the communication circuit is configured to broadcast vehicle information separate from the one or more determined intermediate classifications to one or more other vehicles in a broadcast range of the communication circuit (see Cheng at least [0122] build a wireless message that includes the combined heatmap data, and control the operation of the communication unit 245 to cause the communication unit 245 to broadcast or unicast the wireless message including the combined heatmap data to one or more remote vehicles 124 and [0121] the combined heatmap data describes, for each significant object, (1) the exact position of the significant object in 3D space and (2) the most likely classification for each significant object).
Cheng, Goel, and Herrmann do not teach: the broadcast range depending at least in part on a power of the communication circuit or a type of radio frequency communication of the communication circuit.
However, Chen teaches: the broadcast range depending at least in part on a power of the communication circuit or a type of radio frequency communication of the communication circuit (see Chen at least [0046] Any equipped vehicle 10 within radio communication range of the broadcasting node will receive the probe packet and senses a link).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the inter-vehicle communication and classification system disclosed by Cheng, Goel, and Herrmann to include the power-dependent vehicle-to-vehicle broadcasting range of Chen. One of ordinary skill in the art would have been motivated to make this modification because broadcasting ranges can thusly be adjusted according to circumstances such as vehicle density, as suggested by Chen (see Chen at least [0049] The estimated network density (density of equipped vehicles) is used to set a maximum hop count for a packet relay or the transmission power. Furthermore, the estimated network density can be used to prioritize transmission of packets. For example, in a high density area, only high priority packets are transmitted. In a low-density area, the radio transmission range, e.g., transmitter 220 power can be increased).
Regarding claim 18, Cheng, Goel, and Herrmann disclose: The method of claim 14, comprising: broadcasting vehicle information separate from the determined intermediate classifications to one or more other vehicles in a broadcast range of the communication circuit (see Cheng at least [0122] build a wireless message that includes the combined heatmap data, and control the operation of the communication unit 245 to cause the communication unit 245 to broadcast or unicast the wireless message including the combined heatmap data to one or more remote vehicles 124 and [0121] the combined heatmap data describes, for each significant object, (1) the exact position of the significant object in 3D space and (2) the most likely classification for each significant object).
Cheng, Goel, and Herrmann do not teach: the broadcast range depending at least in part on a power of the communication circuit or a type of radio frequency communication of the communication circuit.
However, Chen teaches: the broadcast range depending at least in part on a power of the communication circuit or a type of radio frequency communication of the communication circuit (see Chen at least [0046] Any equipped vehicle 10 within radio communication range of the broadcasting node will receive the probe packet and senses a link).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the inter-vehicle communication and classification method disclosed by Cheng, Goel, and Herrmann to include the power-dependent vehicle-to-vehicle broadcasting range of Chen. One of ordinary skill in the art would have been motivated to make this modification because broadcasting ranges can thusly be adjusted according to circumstances such as vehicle density, as suggested by Chen (see Chen at least [0049] The estimated network density (density of equipped vehicles) is used to set a maximum hop count for a packet relay or the transmission power. Furthermore, the estimated network density can be used to prioritize transmission of packets. For example, in a high density area, only high priority packets are transmitted. In a low-density area, the radio transmission range, e.g., transmitter 220 power can be increased).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
GB 2558715 A adjusts confidence levels of object classification based on other vehicles’ object classifications.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ELLE ROSE KNUDSON whose telephone number is (703)756-1742. The examiner can normally be reached 1000-1700 ET M-F.
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, Hitesh Patel can be reached at (571) 270-5442. 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.
/ELLE ROSE KNUDSON/Examiner, Art Unit 3667
/ANISS CHAD/Supervisory Patent Examiner, Art Unit 3662