DETAILED ACTION
Notice of Pre-AIA or AIA Status
1. 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
2. This office action is in response to Amendments and Remarks filed on 04/13/2026 for application number 18/667,912 filed on 05/17/2024, in which claims 1-20 were previously presented for examination.
3. Claim(s) 21-25 has/have been added as new, claim(s) 5, 7-8, 14, and 16 has/have been canceled, and claim(s) 1, 3-4, 6, 13, 17, and 19-20 has/have been amended. Accordingly, claim(s) 1-4, 6, 9-13, 15, and 17-25 is/are currently pending.
Priority
4. Acknowledgment is made that applicant has not claimed any foreign or domestic priority.
Prior Art of Record
5. The Examiner has cited particular paragraphs or columns and line numbers in the references applied to the claims above for the convenience of the applicant. Although the specified citations are representative of the teachings of the art and are applied to specific limitations within the individual claim, other passages and figures may apply as well. It is respectfully requested of the applicant in preparing responses, to fully consider the references in their entirety as potentially teaching all or part of the claimed invention, as well as the context of the passage as taught by the prior art or disclosed by the Examiner. The prompt development of a clear issue requires that the replies of the Applicant meet the objections to and rejections of the claims. Applicant should also specifically point out the support for any amendments made to the disclosure (see MPEP §2163.06). Applicant is reminded that the Examiner is entitled to give the Broadest Reasonable Interpretation (BRI) of the language of the claims. Furthermore, the Examiner is not limited to Applicant’s definition which is not specifically set forth in the claims. SEE MPEP 2141.02 [R-07.2015] VI. PRIOR ART MUST BE CONSIDERED IN ITS ENTIRETY, INCLUDING DISCLOSURES THAT TEACH AWAY FROM THE CLAIMS: A prior art reference must be considered in its entirety, i.e., as a whole, including portions that would lead away from the claimed invention. W.L. Gore & Associates, Inc. v. Garlock, Inc., 721 F.2d 1540, 220 USPQ 303 (Fed. Cir. 1983), cert, denied, 469 U.S. 851 (1984). See also MPEP §2123.
Response to Arguments
6. Applicant's arguments filed 04/13/2026 have been fully considered but they are not persuasive.
7. Applicant’s arguments and amendments have been addressed in the new rejection outlined below.
8. Applicant’s arguments with respect to claim(s) 1, 6, and 13 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
9. Applicant argues dependent claim(s) is/are patentable by the virtue of their dependency on one of the independent claims and the additional features recited in the dependent claims.
10. This argument is unpersuasive as each independent claim and dependent claim has been fully rejected and for the reasons given above.
Claim Rejections - 35 USC § 103
11. 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.
12. Claim(s) 1
is/are rejected under 35 U.S.C. 103 as being unpatentable over Rodriguez et al. (US-20230311855-A1) in view of Kotzor et al. (WO-2017103157-A1).
In regard to claim 1
, Rodriguez discloses a system comprising (Rodriguez, in at least [0016], discloses systems and methods related to perception-based parking assistance for autonomous machine applications):
one or more processors (Rodriguez, in at least Fig. 5A, and [0073], discloses the vehicle 500 includes a system(s) on a chip (SoC) 504. The SoC 504 includes CPU(s) 506, GPU(s) 508, processor(s) 510 [i.e., one or more processors], cache(s) 512, accelerator(s) 514, data store(s) 516, and/or other components); and
one or more non-transitory computer-readable media storing instructions executable by the one or more processors, wherein the instructions, when executed, cause the system to perform operations comprising (Rodriguez, in at least [0030], discloses various functions are carried out by a processor executing instructions stored in memory [i.e., storing instructions executable by the one or more processors], such as a tangible memory storage device having a non-transient physical form [i.e., one or more non-transitory computer-readable media]):
receiving sensor data from a sensor associated with an autonomous vehicle (Rodriguez, in at least [0017], discloses a perception-based parking assistance system and corresponding methods parse sensor data captured by on-board sensors [i.e., receiving sensor data from a sensor associated with an autonomous vehicle]);
receiving map data of an environment (Rodriguez, in at least Figs. 5A-5C, and [0058], discloses one or more of the controller(s) 536 receives inputs from an instrument cluster 532 of the vehicle 500 and provide outputs via a human-machine interface (HMI) display 534, an audible annunciator, a loudspeaker, and/or via other components of the vehicle 500. The outputs include information such as vehicle velocity, speed, time, map data (e.g., the HD map 522 of FIG. 5C), location data (e.g., the vehicle’s 500 location, such as on a map), direction, location of other vehicles [i.e., map data of an environment], information about objects and status of objects as perceived by the controller(s) 536, etc. Examiner notes, the map data must be necessarily received by the vehicle before it is used);
determining, based on the sensor data, a dynamic object in the environment (Rodriguez, in at least Figs. 5A-5C, and [0117], discloses a CNN for emergency vehicle detection and identification uses data from microphones 596 [i.e., based on the sensor data] to detect and identify emergency vehicle [i.e., determining … a dynamic object in the environment] sirens);
determining, based on the sensor data and the map data, a representation of the environment, wherein the representation comprises road marking information and velocity information associated with the dynamic object (Rodriguez, in at least Figs. 1, 4, and [0045-0046 & 0087 & 0096 & 0117], discloses the method 400 comprises determining a location of a real-world parking strip relative to an ego-machine and an associated parking rule for the parking strip using a virtual parking strip and one or more virtual parking signs generated based at least in part on one or more detected features in an environment of the ego-machine. The method 400, at block B402, includes detecting, based at least in part on sensor data generated using one or more sensors of an ego-machine, one or more features indicative of parking information associated with at least a portion of a planned path of the ego-machine. The programmable vision accelerator (PVA) is used to perform computer stereo vision. A semi-global matching-based algorithm is used. Many applications for Level 3-5 autonomous driving require motion estimation/stereo matching on-the-fly, such as structure from motion, pedestrian recognition, lane detection, etc. The PVA performs computer stereo vision function on inputs from two monocular cameras. The CNN running on the DLA is trained to identify the relative closing speed of the emergency vehicle [i.e., velocity information associated with the dynamic object]. Examiner notes, lane detection necessarily requires identifying the road marking information. The output of step B402 of method 400, is the representation of the environment which includes the sensor data generated using one or more sensors of an ego-machine, including the location data, direction, location of other vehicles which is the map data of the environment);
inputting the representation comprising the road marking information and the velocity information associated with the dynamic object into a machine learned model (Rodriguez, in at least Figs. 1, 4, and [0020 & 0047], discloses the virtual parking strips is generated for either side of the path based on where the detected features are located, the sequence of their appearance, and/or the information they convey as determined using one or more text recognition algorithms, computer vision algorithms, object detection algorithms, machine learning models, neural networks, and/or the like. The method 400, at block B404, includes computing a geometry of a virtual parking strip based at least in part on the one or more features and tracked motion of the ego-machine. The geometry of the virtual parking strip is computed from a filtered set of features in tandem with the motion of the ego-machine, as discussed herein. The tracked motion of the ego-machine is used to extend the length of the virtual parking strip along the path of travel starting from the location of a first feature until one or more second features are perceived that indicate the end of the parking strip. The locations of the one or more second features is used to generate a second virtual parking sign on the path and an ending boundary of the virtual parking strip. Examiner notes, as portrayed by Fig. 1, the feature data which is derived from the sensor data is an input for identifying and generating the virtual parking strip. That is, the road marking information and the velocity information associated with the dynamic object, which are captured by the sensors, are the input of the machine learned model);
receiving, from the machine learned model, a first output comprising a first probability that a first element of the first output represents a parking location and a second output comprising a second probability that a second element of the second output represents a non-parking location (Rodriguez, in at least Fig. 4, and [0024 & 0048], discloses the perception-based parking assistance system includes or otherwise communicates with an on-board feature classifier that performs the analysis of the sensor data captured by the on-board sensors to identify features that indicate where the starting and ending of a virtual parking strip are defined, and identify what specific parking rule the features convey. The feature classifier is implemented with an artificial intelligence (AI) inference engine or other neural-network technology [i.e., the machine learned model] that is trained to recognize features conveying parking information and to parse the corresponding parking rule from the recognized features. The on-board feature classifier also outputs a confidence level indicating a confidence in the accuracy of the parking rule that it determines from a feature. If a detected feature is partially obstructed or damaged, the feature classifier outputs a parking rule, but also output a confidence level indicating an estimate of the probability that the determined parking rule is correct [i.e., a first output comprising a first probability that a first element of the first output represents a parking location]. The method 400, at block B406, includes associating a parking rule with the virtual parking strip based at least in part on the one or more features. That is, the parking rule is determined at least in part from information obtained from the detected features. The parking rule provides parking related information such as: parking is allowed within the virtual parking strip, stopping is allowed within the virtual parking strip, no parking is allowed within the virtual parking strip, or no stopping is allowed within the virtual parking strip. The parking rule also indicates times, dates, and/or other conditions under which the parking rule is applicable. Where parking or stopping is permitted, the parking rule also indicates if that permission is subject to having a valid permit, such as a disabled parking permit, or if restricted to permitted residents, faculty, or by usage, such as for deliveries only. Examiner notes, generating the confidence level indicating an estimate of the probability that the determined parking rule is correct, necessarily encompasses a second output comprising a second probability that a second element of the second output represents a non-parking location where the second output is generated by subtracting the first probability from 1);
generating dynamic map data based at least in part on the first probability and the second probability, the dynamic map data comprising a first parking spot and a second parking spot (Rodriguez, in at least Fig. 4, and [0049], discloses the method 400, at block B408, includes generating a parking assistance output [i.e., generating dynamic map] indicative of the parking rule and the relative position of the virtual parking strip with respect to the ego-machine. This parking assistance output is used by one or more downstream components of the ego-machine, either to assist an operator in navigating the ego-machine, or as input to perform one or more automated operations for controlling the ego-machine through an environment. Examiner notes, as portrayed by Fig. 4, the output of the step B406, which includes the first and the second probability is used by block B408. As such, the parking assistance output is generated based on the first probability and the second probability, the dynamic map data comprising a first parking spot and a second parking spot, especially when two parking spots are identified);
generating a trajectory based on the first output and the second output (Rodriguez, in at least [0026], discloses the output generated by the perception-based parking assistance system includes a parking assistance output indicative of the parking rule and the relative position of the virtual parking strip with respect to the ego-machine. This parking assistance output is used by one or more downstream components of the ego-machine — such as a world model manager, a path planner, a control component, a localization component, an obstacle avoidance component, an actuation component, and/or the like — to perform one or more operations for controlling the ego-machine through an environment. Examiner notes, the path planner module generates a trajectory, or a planned path that the vehicle uses for navigating the environment. As mentioned above, the parking assistance output, which includes the first output and the second output, is used by the path planner components. That is, generating a trajectory based on the first output and the second output); and
controlling the autonomous vehicle to park in the parking location based on the trajectory and in response to determining to use the first parking spot, wherein controlling the autonomous vehicle to park in the parking location comprises controlling the autonomous vehicle relative to the dynamic object (Rodriguez, in at least Figs. 5A-5C, and [0037 & 0057 & 0117], discloses the downstream navigation components 124 implements automated parking navigation functions that input the parking assistance output from the parking strip processor 102 to identify a valid virtual parking strip for the ego-machine to park or stop within, and then executes its functions to operate the propulsion system 550 and steering system 554 to navigate the ego-machine into a physical location associated with the valid virtual parking strip [i.e., controlling the autonomous vehicle to park in the parking location based on the trajectory]. The controllers(s) 536 receives the parking assistance output from the parking strip processor 102 indicative of the parking rule and the relative position of a virtual parking strip with respect to the ego-machine. With the parking assistance output controller(s) 536 operates the vehicle 500 to navigate into valid parking or stopping parking strips, while avoiding parking strips when parking and/or stopping is not permitted. Once an emergency vehicle is detected, a control program is used to execute an emergency vehicle safety routine, slowing the vehicle, pulling over to the side of the road, parking the vehicle [i.e., controlling the autonomous vehicle], and/or idling the vehicle, with the assistance of ultrasonic sensors 562, until the emergency vehicle(s) passes [i.e., relative to the dynamic object]. Examiner notes, to navigate into a parking, a trajectory must be necessarily generated. Furthermore, finding a parking location according to the limitations above, and parking the vehicle in response to detecting a dynamic object, such as an emergency vehicle, is controlling the autonomous vehicle to park in the parking location based on the trajectory and in response to determining to use the first parking spot, wherein controlling the autonomous vehicle to park in the parking location comprises controlling the autonomous vehicle relative to the dynamic object).
Rodriguez is silent on determining to use the first parking spot based at least in part on the first parking spot being associated with a lower quality level and a higher availability level than the second parking spot;
However, Kotzor teaches determining to use the first parking spot based at least in part on the first parking spot being associated with a lower quality level and a higher availability level than the second parking spot (Kotzor, in at least [0063], teaches a parking space with a high probability of availability [i.e., a higher availability level], but located at a relatively large distance [i.e., a lower quality level] from the destination, is preferred, since the driver is then guaranteed a parking space and can avoid walking or using alternative means of transport);
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Rodriguez in view of Kotzor with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – vehicle systems – and use the high availability parking which is farther form the destination as the first parking spot and control the vehicle to park there and the combination would provide for optimizing a vehicle's parking space search (Kotzor, see at least [0003]).
13. Claim(s) 2
is/are rejected under 35 U.S.C. 103 as being unpatentable over Rodriguez et al. (US-20230311855-A1) in view of Kotzor et al. (WO-2017103157-A1) and further in view of Slattery et al. (US-20240067194-A1).
In regard to claim 2
, Rodriguez, as modified by Kotzor, teaches the system of claim 1, the operations further comprising:
inputting the representation into the machine learned model (Rodriguez, in at least Figs. 1, 4, and [0020 & 0047], discloses the virtual parking strips is generated for either side of the path based on where the detected features are located, the sequence of their appearance, and/or the information they convey as determined using one or more text recognition algorithms, computer vision algorithms, object detection algorithms, machine learning models, neural networks [i.e., the machine learned model], and/or the like. The method 400, at block B404, includes computing a geometry of a virtual parking strip based at least in part on the one or more features and tracked motion of the ego-machine. The geometry of the virtual parking strip is computed from a filtered set of features in tandem with the motion of the ego-machine, as discussed herein. The tracked motion of the ego-machine is used to extend the length of the virtual parking strip along the path of travel starting from the location of a first feature until one or more second features are perceived that indicate the end of the parking strip. The locations of the one or more second features is used to generate a second virtual parking sign on the path and an ending boundary of the virtual parking strip. Examiner notes, as portrayed by Fig. 1, the feature data which is derived from the sensor data is an input for identifying and generating the virtual parking strip. That is, inputting the representation into the machine learned model).
Rodriguez, as modified by Kotzor, is silent on determining that the autonomous vehicle enters a parking destination map area; and
based at least in part on determining that the autonomous vehicle has entered the parking destination map area.
However, Slattery teaches determining that the autonomous vehicle enters a parking destination map area (Slattery, in at least [0029], teaches the machine learning model determines based on concrete pillars and a flat ground surface that a vehicle is in a parking garage [i.e., determining that the autonomous vehicle enters a parking destination map area] and that the vehicle includes an autonomous parking mode for parallel parking or perpendicular parking in an urban parking structure or parking lot); and
based at least in part on determining that the autonomous vehicle has entered the parking destination map area (Slattery, in at least [0029], teaches the machine learning model determines based on concrete pillars and a flat ground surface that a vehicle is in a parking garage [i.e., determining that the autonomous vehicle has entered the parking destination map area] and that the vehicle includes an autonomous parking mode for parallel parking or perpendicular parking in an urban parking structure or parking lot);
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Rodriguez, as already modified by Kotzor, in view of Slattery with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – vehicle systems – and use the machine learning model of Slattery to determine, based on concrete pillars and a flat ground surface, that a vehicle is in a parking garage, and then based on the determination, use the teachings of Rodriguez to input the environment representation into the machine learning model of Rodriguez and the combination would provide for advantageously allowing deployment of improved models, using machine learning, to operate the vehicle and monitor operation of the vehicle (Slattery, see at least [0002]).
14. Claim(s) 3
is/are rejected under 35 U.S.C. 103 as being unpatentable over Rodriguez et al. (US-20230311855-A1) in view of Kotzor et al. (WO-2017103157-A1) and further in view of Tanaka (US-20220212665-A1).
In regard to claim 3
, Rodriguez, as modified by Kotzor, teaches the system of claim 1, the operations further comprising:
generating the dynamic map data to comprise a parking destination area through which the autonomous vehicle is traversing, the dynamic map data comprising temporary data (Rodriguez, in at least Fig. 4, and [0002 & 0049], discloses temporary changes to parking rules are not uncommon, such as when existing parking signs are temporarily covered by local authorities. The method 400, at block B408, includes generating a parking assistance output [i.e., generating dynamic map] indicative of the parking rule and the relative position of the virtual parking strip [i.e., a parking destination area through which the vehicle is traversing] with respect to the ego-machine. This parking assistance output is used by one or more downstream components of the ego-machine, either to assist an operator in navigating the ego-machine, or as input to perform one or more automated operations for controlling the ego-machine through an environment. Examiner notes, if there are temporary changes to parking rules when the dynamic map is generated, the generated dynamic map data comprises temporary data); and
Rodriguez, as modified by Kotzor, is silent on reverting the dynamic map data back to the map data based on at least one of i) a level of change associated with the road marking information and the velocity information being greater than a threshold level of change, or ii) a difference between an initial time at which the dynamic map data is generated and a current time being greater than a threshold difference.
However, Tanaka teaches reverting the dynamic map data back to the map data based on at least one of i) a level of change associated with the road marking information and the velocity information being greater than a threshold level of change, or ii) a difference between an initial time at which the dynamic map data is generated and a current time being greater than a threshold difference (Tanaka, in at least Fig. 1, [0037], teaches the map information stored in the map database 5 is periodically [i.e., a difference between an initial time at which the dynamic map data is generated and a current time being greater than a threshold difference] updated using communication of the vehicle 1 with the outside, SLAM (simultaneous localization and mapping) technology, etc. Examiner notes, updating a map periodically encompasses reverting the dynamic map data back to the map data when the time difference between the periods is greater than a threshold difference).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Rodriguez, as already modified by Kotzor, in view of Tanaka with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – vehicle systems – and update the map periodically and the combination would provide for realizing a vehicle speed suitable for the traffic situation of the surroundings at the time of shifting from automated driving to manual driving (Tanaka, see at least [0017]).
15. Claim(s) 4
is/are rejected under 35 U.S.C. 103 as being unpatentable over Rodriguez et al. (US-20230311855-A1) in view of Kotzor et al. (WO-2017103157-A1) and further in view of Kim et al. (US-20150241880-A1).
In regard to claim 4
, Rodriguez, as modified by Kotzor, teaches the system of claim 1, the operations further comprising:
generating the dynamic map data to comprise a parking destination area through which the autonomous vehicle is traversing, the dynamic map data comprising the map data, the first probability, and the second probability (Rodriguez, in at least Fig. 4, and [0024 & 0049], discloses the perception-based parking assistance system includes or otherwise communicates with an on-board feature classifier that performs the analysis of the sensor data captured by the on-board sensors to identify features that indicate where the starting and ending of a virtual parking strip are defined, and identify what specific parking rule the features convey. The feature classifier is implemented with an artificial intelligence (AI) inference engine or other neural-network technology that is trained to recognize features conveying parking information and to parse the corresponding parking rule from the recognized features. The on-board feature classifier also outputs a confidence level indicating a confidence in the accuracy of the parking rule that it determines from a feature. If a detected feature is partially obstructed or damaged, the feature classifier outputs a parking rule, but also output a confidence level indicating an estimate of the probability that the determined parking rule is correct [i.e., the dynamic map data comprising the map data, the first probability, and the second probability]. The method 400, at block B408, includes generating a parking assistance output [i.e., generating dynamic map] indicative of the parking rule and the relative position of the virtual parking strip with respect to the ego-machine.. Examiner notes, generating the confidence level indicating an estimate of the probability that the determined parking rule is correct, necessarily encompasses a second output comprising a second probability that a second element of the second output represents a non-parking location where the second output is generated by subtracting the first probability from 1. The location where the vehicle is looking for a parking, is the parking destination area through which the autonomous vehicle is traversing. The parking assistance output is generated based on the previous blocks output. That is, generating the dynamic map data to comprise a parking destination area through which the autonomous vehicle is traversing, the dynamic map data comprising the map data, the first probability, and the second probability);
Rodriguez, as modified by Kotzor, is silent on downloading prior dynamic map data associated with a previous autonomous vehicle traversing the environment through the parking destination area; and
updating the dynamic map data based on the prior dynamic map data.
However, Kim teaches downloading prior dynamic map data associated with a previous autonomous vehicle traversing the environment through the parking destination area (Kim, in at least [0042], teaches autonomous vehicles share information among other autonomous vehicles periodically and broadcast information on stopped obstacles. The remote virtual sensor processor of the first autonomous vehicle receives obstacle information (e.g., an obstacle position, an obstacle size, an obstacle attribute, and a time) transmitted by the remote virtual sensor processors of other autonomous vehicles [i.e., downloading prior dynamic map data associated with a previous autonomous vehicle traversing the environment], and delivers the obstacle information to the dynamic map processor when it is determined that there is an obstacle in the stopped obstacle collection area of the first autonomous vehicle. Examiner asserts, an environment encompasses a parking destination area); and
updating the dynamic map data based on the prior dynamic map data (Kim, in at least [0042], teaches the remote virtual sensor processor delivers the obstacle information to the dynamic map processor [i.e., updating the dynamic map data based on the prior dynamic map data] when it is determined that there is an obstacle in the stopped obstacle collection area of the first autonomous vehicle).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Rodriguez, as already modified by Kotzor, in view of Kim with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – vehicle systems – and share dynamic map data and the combination would provide for checking the future driving-related information of the self vehicle and the future driving-related information of the other vehicle, and adjust the driving path of the self vehicle dependent on the future driving-related information of the self vehicle not to overlap a driving path dependent on the future driving-related information of the other vehicle (Kim, see at least [0014]).
16. Claim(s) 6, 13, and 20-25
is/are rejected under 35 U.S.C. 103 as being unpatentable over Rodriguez et al. (US-20230311855-A1) in view of Slattery et al. (US-20240067194-A1).
In regard to claim 6
, Rodriguez discloses a method comprising (Rodriguez, in at least [0016], discloses systems and methods related to perception-based parking assistance for autonomous machine applications):
receiving sensor data and map data associated with an environment, the sensor data associated with a vehicle (Rodriguez, in at least Figs. 5A-5C, and [0017 & 0058], discloses a perception-based parking assistance system and corresponding methods that parse sensor data captured by on-board sensors [i.e., receiving sensor data … associated with an environment, the sensor data associated with a vehicle]. One or more of the controller(s) 536 receives inputs from an instrument cluster 532 of the vehicle 500 and provide outputs (e.g., represented by output data, display data, etc.) via a human-machine interface (HMI) display 534, an audible annunciator, a loudspeaker, and/or via other components of the vehicle 500. The outputs include information such as vehicle velocity, speed, time, map data (e.g., the HD map 522 of FIG. 5C), location data (e.g., the vehicle’s 500 location, such as on a map), direction, location of other vehicles [i.e., map data associated with an environment], information about objects and status of objects as perceived by the controller(s) 536, etc. Examiner notes, the map data must be necessarily received by the vehicle before it is used);
determining, based at least in part on the sensor data, a dynamic object in the environment (Rodriguez, in at least Figs. 5A-5C, and [0117], discloses a CNN for emergency vehicle detection and identification uses data from microphones 596 [i.e., based at least in part on the sensor data] to detect and identify emergency vehicle [i.e., determining … a dynamic object in the environment] sirens);
determining, based at least in part on the map data and the dynamic object, a representation of the environment (Rodriguez, in at least Figs. 1, 4, and [0045-0046 & 0087 & 0096 & 0117], discloses the method 400 comprises determining a location of a real-world parking strip relative to an ego-machine and an associated parking rule for the parking strip using a virtual parking strip and one or more virtual parking signs generated based at least in part on one or more detected features in an environment of the ego-machine. The method 400, at block B402, includes detecting, based at least in part on sensor data generated using one or more sensors of an ego-machine, one or more features indicative of parking information associated with at least a portion of a planned path of the ego-machine. The programmable vision accelerator (PVA) is used to perform computer stereo vision. A semi-global matching-based algorithm is used. Many applications for Level 3-5 autonomous driving require motion estimation/stereo matching on-the-fly, such as structure from motion, pedestrian recognition, lane detection, etc. The PVA performs computer stereo vision function on inputs from two monocular cameras. The CNN running on the DLA is trained to identify the relative closing speed of the emergency vehicle [i.e., based at least in part on … the dynamic object]. Examiner notes, the output of step B402 of method 400, is the representation of the environment which includes the sensor data generated using one or more sensors of an ego-machine, including the location data, direction, location of other vehicles which is the map data of the environment);
inputting the representation into a machine learned model (Rodriguez, in at least Figs. 1, 4, and [0020 & 0047], discloses the virtual parking strips is generated for either side of the path based on where the detected features are located, the sequence of their appearance, and/or the information they convey as determined using one or more text recognition algorithms, computer vision algorithms, object detection algorithms, machine learning models, neural networks, and/or the like. The method 400, at block B404, includes computing a geometry of a virtual parking strip based at least in part on the one or more features and tracked motion of the ego-machine. The geometry of the virtual parking strip is computed from a filtered set of features in tandem with the motion of the ego-machine, as discussed herein. The tracked motion of the ego-machine is used to extend the length of the virtual parking strip along the path of travel starting from the location of a first feature until one or more second features are perceived that indicate the end of the parking strip. The locations of the one or more second features is used to generate a second virtual parking sign on the path and an ending boundary of the virtual parking strip. Examiner notes, as portrayed by Fig. 1, the feature data which is derived from the sensor data is an input for identifying and generating the virtual parking strip. That is, inputting the representation into the machine learned model);
receiving, from the machine learned model, a first output comprising a first probability that a first element of the first output represents a drivable area and a second output comprising a second probability that a second element of the second output represents a parking location (Rodriguez, in at least Fig. 4, and [0024 & 0048], discloses the perception-based parking assistance system includes or otherwise communicates with an on-board feature classifier that performs the analysis of the sensor data captured by the on-board sensors to identify features that indicate where the starting and ending of a virtual parking strip are defined, and identify what specific parking rule the features convey. The feature classifier is implemented with an artificial intelligence (AI) inference engine or other neural-network technology [i.e., the machine learned model] that is trained to recognize features conveying parking information and to parse the corresponding parking rule from the recognized features. The on-board feature classifier also outputs a confidence level indicating a confidence in the accuracy of the parking rule that it determines from a feature. If a detected feature is partially obstructed or damaged, the feature classifier outputs a parking rule, but also output a confidence level indicating an estimate of the probability that the determined parking rule is correct [i.e., a second output comprising a second probability that a second element of the second output represents a parking location]. The method 400, at block B406, includes associating a parking rule with the virtual parking strip based at least in part on the one or more features. That is, the parking rule is determined at least in part from information obtained from the detected features. The parking rule provides parking related information such as: parking is allowed within the virtual parking strip, stopping is allowed within the virtual parking strip, no parking is allowed within the virtual parking strip, or no stopping is allowed within the virtual parking strip. The parking rule also indicates times, dates, and/or other conditions under which the parking rule is applicable. Where parking or stopping is permitted, the parking rule also indicates if that permission is subject to having a valid permit, such as a disabled parking permit, or if restricted to permitted residents, faculty, or by usage, such as for deliveries only. Examiner notes, generating the confidence level indicating an estimate of the probability that the determined parking rule is correct, necessarily encompasses a first output comprising a first probability that a first element of the first output represents a drivable area where the first output is generated by subtracting the second probability from 1); and
determining that the first probability is greater than a threshold probability and the second probability is less than the threshold probability (Rodriguez, in at least Fig. 4, and [0024], discloses the perception-based parking assistance system then applies a confidence threshold such that it will only indicate a valid parking strip exists if the confidence level exceeds the confidence threshold [i.e., determining that the first probability is greater than a threshold probability]. Alternately, the perception-based parking assistance system selects and apply a default parking rule (e.g., “no parking allowed”) to a parking strip when the confidence level falls below the confidence threshold [i.e., the second probability is less than the threshold probability]);
generating a trajectory through the drivable area based at least in part on the first probability being greater than the threshold probability and the second probability being less than the threshold probability (Rodriguez, in at least [0024 & 0026], discloses the perception-based parking assistance system then applies a confidence threshold such that it will only indicate a valid parking strip exists if the confidence level exceeds the confidence threshold [i.e., based at least in part on the first probability being greater than the threshold probability]. Alternately, the perception-based parking assistance system selects and apply a default parking rule (e.g., “no parking allowed”) to a parking strip when the confidence level falls below the confidence threshold [i.e., based at least in part on … the second probability being less than the threshold probability]. The parking assistance output is used by one or more downstream components of the ego-machine — such as a world model manager, a path planner, a control component, a localization component, an obstacle avoidance component, an actuation component, and/or the like — to perform one or more operations for controlling the ego-machine through an environment. Examiner notes, as mentioned above, the parking assistance output, which includes the first output and the second output, is used by the path planner components which encompasses generating a trajectory through the drivable area based at least in part on the first probability being greater than the threshold probability and the second probability being less than the threshold probability), wherein generating the trajectory comprises:
receiving, as first input data, the first probability that the first element of the first output represents the drivable area (Rodriguez, in at least [0024 & 0026], discloses the perception-based parking assistance system includes or otherwise communicates with an on-board feature classifier that performs the analysis of the sensor data captured by the on-board sensors to identify features that indicate where the starting and ending of a virtual parking strip are defined, and identify what specific parking rule the features convey. The feature classifier is implemented with an artificial intelligence (AI) inference engine or other neural-network technology that is trained to recognize features conveying parking information and to parse the corresponding parking rule from the recognized features. The on-board feature classifier also outputs a confidence level indicating a confidence in the accuracy of the parking rule that it determines from a feature. If a detected feature is partially obstructed or damaged, the feature classifier outputs a parking rule, but also output a confidence level indicating an estimate of the probability that the determined parking rule is correct. This parking assistance output is used by one or more downstream components of the ego-machine — such as a world model manager, a path planner, a control component, a localization component, an obstacle avoidance component, an actuation component, and/or the like — to perform one or more operations for controlling the ego-machine through an environment. Examiner notes, generating the confidence level indicating an estimate of the probability that the determined parking rule is correct, necessarily encompasses a first probability that a first element of the first output represents a drivable area where the first output is generated by subtracting the second probability from 1. As mentioned above, the parking assistance output, which includes the first output and the second output, is used by the path planner components which encompasses receiving, as first input data, the first probability that the first element of the first output represents the drivable area);
receiving, as second input data, the second probability that the second element of the second output represents the parking location (Rodriguez, in at least Fig. 4, and [0024 & 0026], discloses the perception-based parking assistance system includes or otherwise communicates with an on-board feature classifier that performs the analysis of the sensor data captured by the on-board sensors to identify features that indicate where the starting and ending of a virtual parking strip are defined, and identify what specific parking rule the features convey. The feature classifier is implemented with an artificial intelligence (AI) inference engine or other neural-network technology that is trained to recognize features conveying parking information and to parse the corresponding parking rule from the recognized features. The on-board feature classifier also outputs a confidence level indicating a confidence in the accuracy of the parking rule that it determines from a feature. If a detected feature is partially obstructed or damaged, the feature classifier outputs a parking rule, but also output a confidence level indicating an estimate of the probability that the determined parking rule is correct [i.e., the second probability that the second element of the second output represents the parking location]. This parking assistance output is used by one or more downstream components of the ego-machine — such as a world model manager, a path planner, a control component, a localization component, an obstacle avoidance component, an actuation component, and/or the like — to perform one or more operations for controlling the ego-machine through an environment. Examiner notes, as mentioned above, the parking assistance output, which includes the first output and the second output, is used by the path planner components which encompasses receiving, as second input data, the second probability that the second element of the second output represents the parking location); and
outputting the trajectory based at least in part on the first input data and the second input data (Rodriguez, in at least [0026], discloses the output generated by the perception-based parking assistance system includes a parking assistance output indicative of the parking rule and the relative position of the virtual parking strip with respect to the ego-machine. This parking assistance output is used by one or more downstream components of the ego-machine — such as a world model manager, a path planner, a control component, a localization component, an obstacle avoidance component, an actuation component, and/or the like — to perform one or more operations for controlling the ego-machine through an environment. Examiner notes, the path planner module generates a trajectory, or a planned path that the vehicle uses for navigating the environment. As mentioned above, the parking assistance output, which includes the first output and the second output, is used by the path planner components. That is, outputting the trajectory based at least in part on the first input data and the second input data); and
controlling the vehicle to park in the parking location based on the trajectory (Rodriguez, in at least Figs. 5A-5C, and [0037 & 0057], discloses the downstream navigation components 124 implements automated parking navigation functions that input the parking assistance output from the parking strip processor 102 to identify a valid virtual parking strip for the ego-machine to park or stop within, and then executes its functions to operate the propulsion system 550 and steering system 554 to navigate the ego-machine into a physical location associated with the valid virtual parking strip [i.e., controlling the vehicle to park in the parking location based on the trajectory]. The controllers(s) 536 receives the parking assistance output from the parking strip processor 102 indicative of the parking rule and the relative position of a virtual parking strip with respect to the ego-machine. With the parking assistance output controller(s) 536 operates the vehicle 500 to navigate into valid parking or stopping parking strips, while avoiding parking strips when parking and/or stopping is not permitted. Examiner notes, to navigate into a valid parking, a trajectory must be necessarily generated).
Rodriguez is silent on determining that the vehicle enters a parking destination map area;
based at least in part on determining that the vehicle has entered the parking destination map area;
However, Slattery teaches determining that the vehicle enters a parking destination map area (Slattery, in at least [0029], teaches the machine learning model determines based on concrete pillars and a flat ground surface that a vehicle is in a parking garage [i.e., determining that the vehicle enters a parking destination map area] and that the vehicle includes an autonomous parking mode for parallel parking or perpendicular parking in an urban parking structure or parking lot); and
based at least in part on determining that the vehicle has entered the parking destination map area (Slattery, in at least [0029], teaches the machine learning model determines based on concrete pillars and a flat ground surface that a vehicle is in a parking garage [i.e., determining that the vehicle has entered the parking destination map area] and that the vehicle includes an autonomous parking mode for parallel parking or perpendicular parking in an urban parking structure or parking lot);
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Rodriguez in view of Slattery with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – vehicle systems – and use the machine learning model of Slattery to determine, based on concrete pillars and a flat ground surface, that a vehicle is in a parking garage, and then based on the determination, use the teachings of Rodriguez to input the environment representation into the machine learning model of Rodriguez and the combination would provide for advantageously allowing deployment of improved models, using machine learning, to operate the vehicle and monitor operation of the vehicle (Slattery, see at least [0002]).
In regard to claim 13
, Rodriguez discloses one or more non-transitory computer-readable media storing instructions executable by one or more processors, wherein the instructions, when executed, cause the one or more processors to perform operations comprising (Rodriguez, in at least [0030], discloses various functions are carried out by a processor [i.e., one or more processors] executing instructions stored in memory [i.e., storing instructions executable by one or more processors], such as a tangible memory storage device having a non-transient physical form [i.e., one or more non-transitory computer-readable media]):
receiving sensor data and map data associated with an environment, wherein the sensor data is received from one or more of a lidar sensor, a radar sensor, or an image sensor associated with a vehicle (Rodriguez, in at least Figs. 5A-5C, and [0017 & 0057-0058], discloses a perception-based parking assistance system and corresponding methods that parse sensor data captured by on-board sensors [i.e., receiving sensor data … associated with an environment, the sensor data associated with a vehicle]. The sensor data is received from, global navigation satellite systems sensor(s) 558 (e.g., Global Positioning System sensor(s)), RADAR sensor(s) 560 [i.e., a radar sensor], ultrasonic sensor(s) 562, LIDAR sensor(s) 564 [i.e., one or more of a lidar sensor], inertial measurement unit (IMU) sensor(s) 566 (e.g., accelerometer(s), gyroscope(s), magnetic compass(es), magnetometer(s), etc.), microphone(s) 596, stereo camera(s) 568, wide-view camera(s) 570 (e.g., fisheye cameras), infrared camera(s) 572, surround camera(s) 574 (e.g., 360 degree cameras), long-range and/or mid-range camera(s) 598 [i.e., an image sensor associated with a vehicle], speed sensor(s) 544 (e.g., for measuring the speed of the vehicle 500), vibration sensor(s) 542, steering sensor(s) 540, brake sensor(s) (e.g., as part of the brake sensor system 546), and/or other sensor types. One or more of the controller(s) 536 receives inputs from an instrument cluster 532 of the vehicle 500 and provide outputs (e.g., represented by output data, display data, etc.) via a human-machine interface (HMI) display 534, an audible annunciator, a loudspeaker, and/or via other components of the vehicle 500. The outputs include information such as vehicle velocity, speed, time, map data (e.g., the HD map 522 of FIG. 5C), location data (e.g., the vehicle’s 500 location, such as on a map), direction, location of other vehicles [i.e., map data associated with an environment], information about objects and status of objects as perceived by the controller(s) 536, etc. Examiner notes, the map data must be necessarily received by the vehicle before it is used);
determining, based at least in part on the sensor data, a dynamic object in the environment (Rodriguez, in at least Figs. 5A-5C, and [0117], discloses a CNN for emergency vehicle detection and identification uses data from microphones 596 [i.e., based at least in part on the sensor data] to detect and identify emergency vehicle [i.e., determining … a dynamic object in the environment] sirens);
determining, based at least in part on the map data and the dynamic object, a representation of the environment (Rodriguez, in at least Figs. 1, 4, and [0045-0046 & 0087 & 0096 & 0117], discloses the method 400 comprises determining a location of a real-world parking strip relative to an ego-machine and an associated parking rule for the parking strip using a virtual parking strip and one or more virtual parking signs generated based at least in part on one or more detected features in an environment of the ego-machine. The method 400, at block B402, includes detecting, based at least in part on sensor data generated using one or more sensors of an ego-machine, one or more features indicative of parking information associated with at least a portion of a planned path of the ego-machine. The programmable vision accelerator (PVA) is used to perform computer stereo vision. A semi-global matching-based algorithm is used. Many applications for Level 3-5 autonomous driving require motion estimation/stereo matching on-the-fly, such as structure from motion, pedestrian recognition, lane detection, etc. The PVA performs computer stereo vision function on inputs from two monocular cameras. The CNN running on the DLA is trained to identify the relative closing speed of the emergency vehicle [i.e., based at least in part on … the dynamic object]. Examiner notes, the output of step B402 of method 400, is the representation of the environment which includes the sensor data generated using one or more sensors of an ego-machine, including the location data, direction, location of other vehicles which is the map data of the environment);
inputting the representation into a machine learned model (Rodriguez, in at least Figs. 1, 4, and [0020 & 0047], discloses the virtual parking strips is generated for either side of the path based on where the detected features are located, the sequence of their appearance, and/or the information they convey as determined using one or more text recognition algorithms, computer vision algorithms, object detection algorithms, machine learning models, neural networks, and/or the like. The method 400, at block B404, includes computing a geometry of a virtual parking strip based at least in part on the one or more features and tracked motion of the ego-machine. The geometry of the virtual parking strip is computed from a filtered set of features in tandem with the motion of the ego-machine, as discussed herein. The tracked motion of the ego-machine is used to extend the length of the virtual parking strip along the path of travel starting from the location of a first feature until one or more second features are perceived that indicate the end of the parking strip. The locations of the one or more second features is used to generate a second virtual parking sign on the path and an ending boundary of the virtual parking strip. Examiner notes, as portrayed by Fig. 1, the feature data which is derived from the sensor data is an input for identifying and generating the virtual parking strip. That is, inputting the representation into the machine learned model);
receiving, from the machine learned model, a first output comprising a first probability that a first element of the first output represents a drivable area and a second output comprising a second probability that a second element of the second output represents a parking location (Rodriguez, in at least Fig. 4, and [0024 & 0048], discloses the perception-based parking assistance system includes or otherwise communicates with an on-board feature classifier that performs the analysis of the sensor data captured by the on-board sensors to identify features that indicate where the starting and ending of a virtual parking strip are defined, and identify what specific parking rule the features convey. The feature classifier is implemented with an artificial intelligence (AI) inference engine or other neural-network technology [i.e., the machine learned model] that is trained to recognize features conveying parking information and to parse the corresponding parking rule from the recognized features. The on-board feature classifier also outputs a confidence level indicating a confidence in the accuracy of the parking rule that it determines from a feature. If a detected feature is partially obstructed or damaged, the feature classifier outputs a parking rule, but also output a confidence level indicating an estimate of the probability that the determined parking rule is correct [i.e., a second output comprising a second probability that a second element of the second output represents a parking location]. The method 400, at block B406, includes associating a parking rule with the virtual parking strip based at least in part on the one or more features. That is, the parking rule is determined at least in part from information obtained from the detected features. The parking rule provides parking related information such as: parking is allowed within the virtual parking strip, stopping is allowed within the virtual parking strip, no parking is allowed within the virtual parking strip, or no stopping is allowed within the virtual parking strip. The parking rule also indicates times, dates, and/or other conditions under which the parking rule is applicable. Where parking or stopping is permitted, the parking rule also indicates if that permission is subject to having a valid permit, such as a disabled parking permit, or if restricted to permitted residents, faculty, or by usage, such as for deliveries only. Examiner notes, generating the confidence level indicating an estimate of the probability that the determined parking rule is correct, necessarily encompasses a first output comprising a first probability that a first element of the first output represents a drivable area where the first output is generated by subtracting the second probability from 1);
generating a trajectory based at least in part on the first output and the second output (Rodriguez, in at least Fig. 4, and [0026], discloses the parking assistance output is used by one or more downstream components of the ego-machine — such as a world model manager, a path planner, a control component, a localization component, an obstacle avoidance component, an actuation component, and/or the like — to perform one or more operations for controlling the ego-machine through an environment. Examiner notes, as mentioned above, the parking assistance output, which includes the first output and the second output, is used by the path planner components which encompasses receiving, as second input data, the second probability that the second element of the second output represents the parking location), wherein generating the trajectory comprises:
receiving, as first input data, the first probability that the first element of the first output represents the drivable area (Rodriguez, in at least [0024 & 0026], discloses the perception-based parking assistance system includes or otherwise communicates with an on-board feature classifier that performs the analysis of the sensor data captured by the on-board sensors to identify features that indicate where the starting and ending of a virtual parking strip are defined, and identify what specific parking rule the features convey. The feature classifier is implemented with an artificial intelligence (AI) inference engine or other neural-network technology that is trained to recognize features conveying parking information and to parse the corresponding parking rule from the recognized features. The on-board feature classifier also outputs a confidence level indicating a confidence in the accuracy of the parking rule that it determines from a feature. If a detected feature is partially obstructed or damaged, the feature classifier outputs a parking rule, but also output a confidence level indicating an estimate of the probability that the determined parking rule is correct. This parking assistance output is used by one or more downstream components of the ego-machine — such as a world model manager, a path planner, a control component, a localization component, an obstacle avoidance component, an actuation component, and/or the like — to perform one or more operations for controlling the ego-machine through an environment. Examiner notes, generating the confidence level indicating an estimate of the probability that the determined parking rule is correct, necessarily encompasses a first probability that a first element of the first output represents a drivable area where the first output is generated by subtracting the second probability from 1. As mentioned above, the parking assistance output, which includes the first output and the second output, is used by the path planner components which encompasses receiving, as first input data, the first probability that the first element of the first output represents the drivable area);
receiving, as second input data, the second probability that the second element of the second output represents the parking location (Rodriguez, in at least [0024 & 0026], discloses the perception-based parking assistance system includes or otherwise communicates with an on-board feature classifier that performs the analysis of the sensor data captured by the on-board sensors to identify features that indicate where the starting and ending of a virtual parking strip are defined, and identify what specific parking rule the features convey. The feature classifier is implemented with an artificial intelligence (AI) inference engine or other neural-network technology that is trained to recognize features conveying parking information and to parse the corresponding parking rule from the recognized features. The on-board feature classifier also outputs a confidence level indicating a confidence in the accuracy of the parking rule that it determines from a feature. If a detected feature is partially obstructed or damaged, the feature classifier outputs a parking rule, but also output a confidence level indicating an estimate of the probability that the determined parking rule is correct [i.e., the second probability that the second element of the second output represents the parking location]. This parking assistance output is used by one or more downstream components of the ego-machine — such as a world model manager, a path planner, a control component, a localization component, an obstacle avoidance component, an actuation component, and/or the like — to perform one or more operations for controlling the ego-machine through an environment. Examiner notes, as mentioned above, the parking assistance output, which includes the first output and the second output, is used by the path planner components which encompasses receiving, as second input data, the second probability that the second element of the second output represents the parking location); and
outputting the trajectory based at least in part on the first input data and the second input data (Rodriguez, in at least [0026], discloses the output generated by the perception-based parking assistance system includes a parking assistance output indicative of the parking rule and the relative position of the virtual parking strip with respect to the ego-machine. This parking assistance output is used by one or more downstream components of the ego-machine — such as a world model manager, a path planner, a control component, a localization component, an obstacle avoidance component, an actuation component, and/or the like — to perform one or more operations for controlling the ego-machine through an environment. Examiner notes, the path planner module generates a trajectory, or a planned path that the vehicle uses for navigating the environment. As mentioned above, the parking assistance output, which includes the first output and the second output, is used by the path planner components. That is, outputting the trajectory based at least in part on the first input data and the second input data); and
controlling the vehicle to park in the parking location traverse the environment based on the trajectory (Rodriguez, in at least Figs. 5A-5C, and [0037 & 0057], discloses the downstream navigation components 124 implements automated parking navigation functions that input the parking assistance output from the parking strip processor 102 to identify a valid virtual parking strip for the ego-machine to park or stop within, and then executes its functions to operate the propulsion system 550 and steering system 554 to navigate the ego-machine into a physical location associated with the valid virtual parking strip [i.e., controlling the vehicle to park in the parking location traverse the environment based on the trajectory]. The controllers(s) 536 receives the parking assistance output from the parking strip processor 102 indicative of the parking rule and the relative position of a virtual parking strip with respect to the ego-machine. With the parking assistance output controller(s) 536 operates the vehicle 500 to navigate into valid parking or stopping parking strips, while avoiding parking strips when parking and/or stopping is not permitted. Examiner notes, to navigate into a valid parking, a trajectory must be necessarily generated).
Rodriguez is silent on determining that the vehicle enters a parking destination map area;
based at least in part on determining that the vehicle has entered the parking destination map area;
However, Slattery teaches determining that the vehicle enters a parking destination map area (Slattery, in at least [0029], teaches the machine learning model determines based on concrete pillars and a flat ground surface that a vehicle is in a parking garage [i.e., determining that the vehicle enters a parking destination map area] and that the vehicle includes an autonomous parking mode for parallel parking or perpendicular parking in an urban parking structure or parking lot); and
based at least in part on determining that the vehicle has entered the parking destination map area (Slattery, in at least [0029], teaches the machine learning model determines based on concrete pillars and a flat ground surface that a vehicle is in a parking garage [i.e., determining that the vehicle enters a parking destination map area] and that the vehicle includes an autonomous parking mode for parallel parking or perpendicular parking in an urban parking structure or parking lot);
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Rodriguez in view of Slattery with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – vehicle systems – and use the machine learning model of Slattery to determine, based on concrete pillars and a flat ground surface, that a vehicle is in a parking garage, and then based on the determination, use the teachings of Rodriguez to input the environment representation into the machine learning model of Rodriguez and the combination would provide for advantageously allowing deployment of improved models, using machine learning, to operate the vehicle and monitor operation of the vehicle (Slattery, see at least [0002]).
In regard to claim 20
, Rodriguez, as modified by Slattery, teaches the one or more non-transitory computer-readable media of claim 13, wherein the instructions, when executed, cause the one or more processors to perform further operations comprising (Rodriguez, in at least [0030], discloses various functions are carried out by a processor executing instructions stored in memory, such as a tangible memory storage device having a non-transient physical form):
receiving from the machine learned model, a boundary line associated with a parking location identified by the machine learned model (Rodriguez, in at least Fig. 3, and [0020-0040], discloses the virtual parking strips is generated for either side of the path based on where the detected features are located, the sequence of their appearance, and/or the information they convey as determined using one or more text recognition algorithms, computer vision algorithms, object detection algorithms, machine learning models, neural networks [i.e., identified by the machine learned model], and/or the like is permitted to stop or park. Each of the virtual parking strips is shown to include virtual parking signs indicating their respective starting and ending boundaries [i.e., a boundary line associated with the parking location]. These virtual parking signs would be generated by the virtual parking strip generator 210 in response to the parking information detected from feature data 104); and
determining the trajectory based at least in part on the boundary line (Rodriguez, in at least [0026], discloses the output generated by the perception-based parking assistance system includes a parking assistance output indicative of the parking rule and the relative position of the virtual parking strip with respect to the ego-machine. This parking assistance output is used by one or more downstream components of the ego-machine — such as a world model manager, a path planner, a control component, a localization component, an obstacle avoidance component, an actuation component, and/or the like — to perform one or more operations for controlling the ego-machine through an environment. Examiner notes, the path planner module generates a trajectory, or a planned path that the vehicle uses for navigating the environment. Accordingly, parking the vehicle in one of the identified parking locations encompasses determining the trajectory based at least in part on the boundary line).
In regard to claim 21
, Rodriguez, as modified by Slattery, teaches the one or more non-transitory computer-readable media of claim 13, wherein the vehicle is an autonomous vehicle (Rodriguez, in at least [0017], discloses perception-based generation of parking information for use by an autonomous machine. More specifically, the systems and methods assist an autonomous machine in parsing information obtained from their environment to determine locations where the autonomous vehicle [i.e., wherein the vehicle is an autonomous vehicle] is permitted to stop or park).
In regard to claim 22
, Rodriguez, as modified by Slattery, teaches the one or more non-transitory computer-readable media of claim 13, wherein the parking location is output by the machine learned model (Rodriguez, in at least [0020], discloses the virtual parking strips is generated for either side of the path based on where the detected features are located, the sequence of their appearance, and/or the information they convey as determined using one or more text recognition algorithms, computer vision algorithms, object detection algorithms, machine learning models, neural networks [i.e., wherein the parking location is output by the machine learned model], and/or the like is permitted to stop or park).
In regard to claim 23
, Rodriguez, as modified by Slattery, teaches the method of claim 6, further comprising:
receiving from the machine learned model, a boundary line associated with the parking location identified by the machine learned model (Rodriguez, in at least Fig. 3, and [0020-0040], discloses the virtual parking strips is generated for either side of the path based on where the detected features are located, the sequence of their appearance, and/or the information they convey as determined using one or more text recognition algorithms, computer vision algorithms, object detection algorithms, machine learning models, neural networks [i.e., identified by the machine learned model], and/or the like is permitted to stop or park. Each of the virtual parking strips is shown to include virtual parking signs indicating their respective starting and ending boundaries [i.e., a boundary line associated with the parking location]. These virtual parking signs would be generated by the virtual parking strip generator 210 in response to the parking information detected from feature data 104); and
determining the trajectory based at least in part on the boundary line (Rodriguez, in at least [0026], discloses the output generated by the perception-based parking assistance system includes a parking assistance output indicative of the parking rule and the relative position of the virtual parking strip with respect to the ego-machine. This parking assistance output is used by one or more downstream components of the ego-machine — such as a world model manager, a path planner, a control component, a localization component, an obstacle avoidance component, an actuation component, and/or the like — to perform one or more operations for controlling the ego-machine through an environment. Examiner notes, the path planner module generates a trajectory, or a planned path that the vehicle uses for navigating the environment. Accordingly, to park the vehicle in one of the identified parking locations encompasses determining the trajectory based at least in part on the boundary line).
In regard to claim 24
, Rodriguez, as modified by Slattery, teaches the method of claim 6, wherein the vehicle is an autonomous vehicle (Rodriguez, in at least [0017], discloses perception-based generation of parking information for use by an autonomous machine. More specifically, the systems and methods assist an autonomous machine in parsing information obtained from their environment to determine locations where the autonomous vehicle [i.e., wherein the vehicle is an autonomous vehicle] is permitted to stop or park).
In regard to claim 25
, Rodriguez, as modified by Slattery, teaches the method of claim 6, wherein the sensor data is received from one or more of a lidar sensor, a radar sensor, or an image sensor (Rodriguez, in at least Figs. 5A-5C, and [0058], discloses the sensor data is received from, global navigation satellite systems sensor(s) 558 (e.g., Global Positioning System sensor(s)), RADAR sensor(s) 560 [i.e., a radar sensor], ultrasonic sensor(s) 562, LIDAR sensor(s) 564 [i.e., one or more of a lidar sensor], inertial measurement unit (IMU) sensor(s) 566 (e.g., accelerometer(s), gyroscope(s), magnetic compass(es), magnetometer(s), etc.), microphone(s) 596, stereo camera(s) 568, wide-view camera(s) 570 (e.g., fisheye cameras), infrared camera(s) 572, surround camera(s) 574 (e.g., 360 degree cameras), long-range and/or mid-range camera(s) 598 [i.e., an image sensor associated with a vehicle], speed sensor(s) 544 (e.g., for measuring the speed of the vehicle 500), vibration sensor(s) 542, steering sensor(s) 540, brake sensor(s) (e.g., as part of the brake sensor system 546), and/or other sensor types).
17. Claim(s) 9
is/are rejected under 35 U.S.C. 103 as being unpatentable over Rodriguez et al. (US-20230311855-A1) in view of Slattery et al. (US-20240067194-A1) and further in view of Chan (US-20180033302-A1).
In regard to claim 9
, Rodriguez, as modified by Slattery, teaches the method of claim 6, accordingly the rejection of claim 6 is incorporated.
Rodriguez, as modified by Slattery, is silent on further comprising:
applying a threshold to the first output to identify first candidate parking locations in the environment;
applying a size filter to the first candidate parking locations; and
determining, based on the size filter and the first candidate parking locations, second candidate parking locations in the environment.
However, Chan teaches further comprising:
applying a threshold to the first output to identify first candidate parking locations in the environment (Chan, in at least [0029], teaches a size threshold is provided for the vehicle dimensions. The size threshold is used to account for extra space needed to get into and out of a parking spot [i.e., applying a threshold to the first output to identify first candidate parking locations]);
applying a size filter to the first candidate parking locations (Chan, in at least [0029], teaches the actual dimensions for each of the potential parking spots is compared to the size threshold in order to filter out [i.e., applying a size filter to the first candidate parking locations] unsuitable spots from the plurality of potential parking spots); and
determining, based on the size filter and the first candidate parking locations, second candidate parking locations in the environment (Chan, in at least [0029], teaches the unsuitable spots are smaller than a minimum allowable size for the suitable parking spots [i.e., ] and therefore would not be a viable parking option. As mentioned above, a parking space that is smaller than the vehicle dimensions will be rejected. That is, a parking spot will be selected that can accommodate the vehicle dimensions which is determining, based on the size filter and the first candidate parking locations, second candidate parking locations in the environment).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Rodriguez, as already modified by Slattery, in view of Chan with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – vehicle systems – and use the dimensions of a vehicle for finding a parking and the combination would provide for detecting potential hazards, as well as areas in which parking is not authorized (Chan, see at least [0004]).
18. Claim(s) 10
is/are rejected under 35 U.S.C. 103 as being unpatentable over Rodriguez et al. (US-20230311855-A1) in view of Slattery et al. (US-20240067194-A1) and further in view of Frankfurth (US-20200211371-A1).
In regard to claim 10
, Rodriguez, as modified by Slattery, teaches the method of claim 6, further comprising:
determining, by the machine learned model, the parking location (Rodriguez, in at least Figs. 1, 4, and [0020], discloses the virtual parking strips is generated for either side of the path based on where the detected features are located, the sequence of their appearance, and/or the information they convey as determined using one or more text recognition algorithms, computer vision algorithms, object detection algorithms, machine learning models, neural networks, and/or the like. Examiner notes, generating virtual parking strips encompasses determining, by the machine learned model, a parking location),
in response to determining the parking location, controlling the vehicle to park in the parking location (Rodriguez, in at least Figs. 5A-5C, and [0037 & 0057], discloses the downstream navigation components 124 implements automated parking navigation functions that input the parking assistance output from the parking strip processor 102 to identify a valid virtual parking strip for the ego-machine to park or stop within, and then executes its functions to operate the propulsion system 550 and steering system 554 to navigate the ego-machine into a physical location associated with the valid virtual parking strip [i.e., controlling the vehicle to park in the parking location]. The controllers(s) 536 receives the parking assistance output from the parking strip processor 102 indicative of the parking rule and the relative position of a virtual parking strip with respect to the ego-machine. With the parking assistance output controller(s) 536 operates the vehicle 500 to navigate into valid parking or stopping parking strips, while avoiding parking strips when parking and/or stopping is not permitted. Examiner notes, controlling the vehicle to park in the parking location is based on generating the virtual parking strip. That is, controlling the vehicle to park in the parking location in response to determining the parking location).
Rodriguez, as modified by Slattery, is silent on determining that a speed of the dynamic object is less than a threshold speed; and
determining, … the parking location, based at least in part on the speed of the dynamic object being below the threshold speed;
However, Frankfurth teaches determining that a speed of a dynamic object is less than a threshold speed (Frankfurth, in at least [0030], teaches a parking space is identified when one of the movement patterns comprises identifying objects [i.e., dynamic object] repeatedly moving back and forth in a common area at a speed that is less than a predetermined speed [i.e., the speed of the dynamic object being below the threshold speed], e.g. forward and reverse maneuvering in a small defined area); and
determining, … the parking location, based at least in part on the speed of the dynamic object being below the threshold speed (Frankfurth, in at least [0030], teaches a parking space is identified [i.e., determining, … the parking location] when one of the movement patterns comprises identifying objects [i.e., dynamic object] repeatedly moving back and forth in a common area at a speed that is less than a predetermined speed [i.e., the speed of the dynamic object being below the threshold speed], e.g. forward and reverse maneuvering in a small defined area);
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Rodriguez, as already modified by Slattery, in view of Frankfurth with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – vehicle systems – and identify a parking space by detecting an object moving ack and forth at a speed that is less than a predetermined speed and the combination would provide for identifying movement patterns in an intersection and creating a map of the intersection based on learned movement patterns (Frankfurth, see at least [0002]).
19. Claim(s) 11
is/are rejected under 35 U.S.C. 103 as being unpatentable over Rodriguez et al. (US-20230311855-A1) in view of Slattery et al. (US-20240067194-A1) and further in view of Wen et al. (US-20230294669-A1).
In regard to claim 11
, Rodriguez, as modified by Slattery, teaches the method of claim 6, accordingly the rejection of claim 6 is incorporated.
Rodriguez, as modified by Slattery, is silent on wherein the parking location is output by the machine learned model, and the parking location is different than an initial parking location indicated by the map data.
However, Wen teaches wherein the parking location is output by the machine learned model, and the parking location is different than an initial parking location indicated by the map data (Wen, in at least [0041], teaches when the map information is available, the reliability of the image information is verified through the map information, whether the image information is available and stored is determined, and the availability of the image information is updated (substantially) in real time because the map information and the image information are updated (substantially) in real time. Examiner notes when the map is updated, it means the parking location in the image is different than an initial parking location indicated by the map data).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Rodriguez, as already modified by Slattery, in view of Wen with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – navigation and vehicle control system – and update the map data when the location of the parking in the picture is different from the location of the parking based on the map data and the combination would provide for the safety of the vehicle and the driver (Wen, see at least [0015]).
20. Claim(s) 12
is/are rejected under 35 U.S.C. 103 as being unpatentable over Rodriguez et al. (US-20230311855-A1) in view of Slattery et al. (US-20240067194-A1) and further in view of Kim et al. (US-20150241880-A1).
In regard to claim 12
, Rodriguez, as modified by Slattery, teaches the method of claim 6, further comprising:
generating dynamic map data comprising a parking destination area through which the vehicle is traversing, the dynamic map data comprising the map data, the first probability, and the second probability (Rodriguez, in at least Fig. 4, and [0024 & 0049], discloses the perception-based parking assistance system includes or otherwise communicates with an on-board feature classifier that performs the analysis of the sensor data captured by the on-board sensors to identify features that indicate where the starting and ending of a virtual parking strip are defined, and identify what specific parking rule the features convey. The feature classifier is implemented with an artificial intelligence (AI) inference engine or other neural-network technology that is trained to recognize features conveying parking information and to parse the corresponding parking rule from the recognized features. The on-board feature classifier also outputs a confidence level indicating a confidence in the accuracy of the parking rule that it determines from a feature. If a detected feature is partially obstructed or damaged, the feature classifier outputs a parking rule, but also output a confidence level indicating an estimate of the probability that the determined parking rule is correct [i.e., the dynamic map data comprising the map data, the first probability, and the second probability]. The method 400, at block B408, includes generating a parking assistance output [i.e., generating dynamic map] indicative of the parking rule and the relative position of the virtual parking strip with respect to the ego-machine.. Examiner notes, generating the confidence level indicating an estimate of the probability that the determined parking rule is correct, necessarily encompasses a second output comprising a second probability that a second element of the second output represents a non-parking location where the second output is generated by subtracting the first probability from 1. The location where the vehicle is looking for a parking, is the parking destination area through which the autonomous vehicle is traversing. The parking assistance output is generated based on the previous blocks output. That is, generating the dynamic map data to comprise a parking destination area through which the autonomous vehicle is traversing, the dynamic map data comprising the map data, the first probability, and the second probability);
Rodriguez, as modified by Slattery, is silent on downloading prior dynamic map data associated with a previous vehicle traversing the environment through the parking destination area; and
updating the dynamic map data based on the prior dynamic map data.
However, Kim teaches downloading prior dynamic map data associated with a previous vehicle traversing the environment through the parking destination area (Kim, in at least [0042], teaches autonomous vehicles share information among other autonomous vehicles periodically and broadcast information on stopped obstacles. The remote virtual sensor processor of the first autonomous vehicle receives obstacle information (e.g., an obstacle position, an obstacle size, an obstacle attribute, and a time) [i.e., downloading prior dynamic map data associated with a previous vehicle traversing the environment] transmitted by the remote virtual sensor processors of other autonomous vehicles, and delivers the obstacle information to the dynamic map processor when it is determined that there is an obstacle in the stopped obstacle collection area of the first autonomous vehicle. Examiner asserts, an environment encompasses a parking destination area); and
updating the dynamic map data based on the prior dynamic map data (Kim, in at least [0042], teaches The remote virtual sensor processor delivers the obstacle information to the dynamic map processor [i.e., updating the dynamic map data based on the prior dynamic map data] when it is determined that there is an obstacle in the stopped obstacle collection area of the first autonomous vehicle).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Rodriguez, as already modified by Slattery, in view of Kim with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – vehicle systems – and share dynamic map data and the combination would provide for checking the future driving-related information of the self vehicle and the future driving-related information of the other vehicle, and adjust the driving path of the self vehicle dependent on the future driving-related information of the self vehicle not to overlap a driving path dependent on the future driving-related information of the other vehicle (Kim, see at least [0014]).
21. Claim(s) 15
is/are rejected under 35 U.S.C. 103 as being unpatentable over Rodriguez et al. (US-20230311855-A1) in view of Slattery et al. (US-20240067194-A1) and further in view of Parikh (US-20220207275-A1).
In regard to claim 15
, Rodriguez, as modified by Slattery, teaches the one or more non-transitory computer-readable media of claim 13, accordingly the rejection of claim 13 is incorporated.
Rodriguez, as modified by Slattery,, is silent on wherein the representation comprises a top-down representation, and the top-down representation comprises a multi-channel image or polylines.
However, Parikh teaches wherein the representation comprises a top-down representation, and the top-down representation comprises a multi-channel image or polylines (Parikh, in at least [0011], teaches environment data is represented as multi-channel image data [i.e., a multi-channel image], which represents a top-down view of the environment [i.e., a top-down representation]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Rodriguez, as already modified by Slattery, in view of Parikh with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – vehicle systems – and use a top-down view of the environment and the combination would provide for determining a classification probability of an object in an environment (Parikh, see at least [0009]).
22. Claim(s) 17
is/are rejected under 35 U.S.C. 103 as being unpatentable over Rodriguez et al. (US-20230311855-A1) in view of Slattery et al. (US-20240067194-A1) and further in view of Houston (US-20210197720-A1).
_In regard to claim 17
, Rodriguez, as modified by Slattery, teaches the one or more non-transitory computer-readable media of claim 13, wherein the instructions, when executed, cause the one or more processors to perform further operations comprising (Rodriguez, in at least [0030], discloses various functions are carried out by a processor executing instructions stored in memory, such as a tangible memory storage device having a non-transient physical form):
determining perception data that comprises previous road marking information and velocity information, the previous road marking information and velocity information comprising velocity data associated with objects in the environment (Rodriguez, in at least Figs. 1, 2, 4, and [0035 & 0045-0046 & 0087 & 0096 & 0117], discloses as virtual parking strips are generated by the virtual parking strip generator 210, information about the virtual parking strips may be stored to a virtual park strip memory 218. Virtual park strip memory 218 includes a history of previously computed virtual parking signs [i.e., perception data] generated for each virtual parking strip [i.e., previous road marking information and velocity information], along with their relative distance from the ego-machine and/or from each other. The virtual parking strip generator 210 further updates such relative distances for virtual parking signs stored in the virtual park strip memory 218 based on the tracked motion data 112. The method 400 comprises determining a location of a real-world parking strip relative to an ego-machine and an associated parking rule for the parking strip using a virtual parking strip and one or more virtual parking signs generated based at least in part on one or more detected features in an environment of the ego-machine. The method 400, at block B402, includes detecting, based at least in part on sensor data generated using one or more sensors of an ego-machine, one or more features indicative of parking information associated with at least a portion of a planned path of the ego-machine. The programmable vision accelerator (PVA) is used to perform computer stereo vision. A semi-global matching-based algorithm is used. Many applications for Level 3-5 autonomous driving require motion estimation/stereo matching on-the-fly, such as structure from motion, pedestrian recognition, lane detection, etc. The PVA performs computer stereo vision function on inputs from two monocular cameras. The CNN running on the DLA is trained to identify the relative closing speed of the emergency vehicle [i.e., velocity information associated with the dynamic object]. Examiner notes, lane detection necessarily requires identifying the road marking information. The output of step B402 of method 400, is the representation of the environment which includes the sensor data generated using one or more sensors of an ego-machine, including the location data, direction, location of other vehicles which is the map data of the environment. Examiner notes, As mentioned above, virtual park strip memory 218 includes the history of previously computed virtual parking signs. As such, the perception data includes all the marking information and velocity information associated with the dynamic object of the time that the data has been collected. That is, determining perception data that comprises previous road marking information and velocity information, the previous road marking information and velocity information comprising velocity data associated with objects in the environment); and
.
Rodriguez, as modified by Slattery, is silent on training the machine learned model based at least in part on the perception data.
However, Houston teaches training the machine learned model based at least in part on the perception data (Houston, in at least Fig. 1, [0037], teaches the vehicle system trains the machine-learning model [i.e., training the machine learned model] by updating the machine-learning model to reflect that the historical perception data [i.e., based at least in part on the perception data] for the time in the past is associated with the retrieved risk score).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Rodriguez, as already modified by Slattery, in view of Houston with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – vehicle systems – and train the machine learning model by using the historical perception data and the combination would provide for increased safety and improved estimated times of arrival (Houston, see at least [0017]).
23. Claim(s) 18
is/are rejected under 35 U.S.C. 103 as being unpatentable over Rodriguez et al. (US-20230311855-A1) in view of Slattery et al. (US-20240067194-A1) and further in view of Tanaka (US-20220212665-A1).
In regard to claim 18
, Rodriguez, as modified by Slattery, teaches the one or more non-transitory computer-readable media of claim 13, wherein the instructions, when executed, cause the one or more processors to perform further operations comprising (Rodriguez, in at least [0030], discloses various functions are carried out by a processor executing instructions stored in memory, such as a tangible memory storage device having a non-transient physical form):
generating dynamic map data comprising a parking destination area through which the vehicle is traversing, the dynamic map data comprising temporary data (Rodriguez, in at least Fig. 4, and [0002 & 0049], discloses temporary changes to parking rules are not uncommon, such as when existing parking signs are temporarily covered by local authorities. The method 400, at block B408, includes generating a parking assistance output [i.e., generating dynamic map] indicative of the parking rule and the relative position of the virtual parking strip [i.e., a parking destination area through which the vehicle is traversing] with respect to the ego-machine. This parking assistance output is used by one or more downstream components of the ego-machine, either to assist an operator in navigating the ego-machine, or as input to perform one or more automated operations for controlling the ego-machine through an environment. Examiner notes, if there are temporary changes to parking rules when the dynamic map is generated, the generated dynamic map data comprises temporary data); and
Rodriguez, as modified by Slattery, is silent on reverting the dynamic map data back to the map data based on at least one of i) a level of change associated with at least road marking information and velocity information being greater than a threshold level of change, or ii) a difference between an initial time at which the dynamic map data is generated and a current time being greater than a threshold difference.
However, Tanaka teaches reverting the dynamic map data back to the map data based on at least one of i) a level of change associated with the at least road marking and velocity information being greater than a threshold level of change, or ii) a difference between an initial time at which the dynamic map data is generated and a current time being greater than a threshold difference (Tanaka, in at least Fig. 1, [0037], teaches the map information stored in the map database 5 is periodically [i.e., a difference between an initial time at which the dynamic map data is generated and a current time being greater than a threshold difference] updated using communication of the vehicle 1 with the outside, SLAM (simultaneous localization and mapping) technology, etc. Examiner notes, updating a map periodically encompasses reverting the dynamic map data back to the map data when the time difference between the periods is greater than a threshold difference).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Rodriguez, as already modified by Slattery, in view of Tanaka with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – vehicle systems – and update the map periodically and the combination would provide for realizing a vehicle speed suitable for the traffic situation of the surroundings at the time of shifting from automated driving to manual driving (Tanaka, see at least [0017]).
24. Claim(s) 19
is/are rejected under 35 U.S.C. 103 as being unpatentable over Rodriguez et al. (US-20230311855-A1) in view of Slattery et al. (US-20240067194-A1) and further in view of Goyal et al. (US-20230074387-A1).
In regard to claim 19
, Rodriguez, as modified by Slattery, teaches the one or more non-transitory computer-readable media of claim 13, wherein the instructions, when executed, cause the one or more processors to perform further operations comprising (Rodriguez, in at least [0030], discloses various functions are carried out by a processor executing instructions stored in memory, such as a tangible memory storage device having a non-transient physical form):
Rodriguez, as modified by Slattery, is silent on determining log data of another vehicle under control of a driver and traversing the parking destination map area; and
training the machine learned model based at least in part on the log data.
However, Goyal teaches determining log data of another vehicle under control of a driver and traversing the parking destination map area (Goyal, in at least Figs. 2, 4, and [0042 & 0064], teaches a driveway or parking area is identified by a geolocated polygon marking out the boundaries of the driveway or parking area such as polygons 280, 282, 284. Driveway and parking area boundaries are identified by human labelers [i.e., determining log data of another vehicle under control of a driver and traversing the parking destination map area] or by using such labeled data to train machine learning models to identify polygons for driveways using input map information. The storage system 450 stores log data [i.e., log data] which includes data generated by the various systems of a vehicle, such as autonomous vehicle 100, while the vehicle is being operated in a manual driving mode); and
training the machine learned model based at least in part on the log data (Goyal, in at least Figs. 2, and [0042], teaches a driveway or parking area is identified by a geolocated polygon marking out the boundaries of the driveway or parking area such as polygons 280, 282, 284. Driveway and parking area boundaries are identified by human labelers or by using such labeled data to train machine learning models [i.e., training the machine learned model based at least in part on the log data] to identify polygons for driveways using input map information).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Rodriguez, as already modified by Slattery, in view of Goyal with a reasonable expectation of success, as all inventions are directed to the same field of endeavor – vehicle systems – and driveway and parking area boundaries are identified by human labelers and the labels are used for training a model and the combination would provide for improving the accuracy of predictions, for entry points and off road areas (Goyal, see at least [0002]).
Conclusion
25. The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Beaurepaire et al. (US-20230400312-A1) teaches building a machine learning model to map and predict vehicular wait events using map data and/or vehicle sensor data.
Zilberman et al. (US-20230100851-A1) teaches a system capable of mapping an indoor and/or underground facility (e.g., a parking facility) using multi-modal trajectories collected using mobile devices.
Gray (US-20180373263-A1) teaches the identification of drivable space in each received image frame is performed by applying one or more machine-learned models.
Chong et al. (US-20230069215-A1) teaches applying machine learning to the image data and detecting a drivable area in a parking facility based on the classification score which is a probability value.
26. Any inquiry concerning this communication or earlier communications from the examiner should be directed to Preston J Miller whose telephone number is (703)756-1582. The examiner can normally be reached Monday through Friday 7:30 AM - 4:30 PM EST.
27. 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.
28. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Ramya P Burgess can be reached at (571) 272-6011. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
29. 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.
/P.J.M./Examiner, Art Unit 3661
/Tarek Elarabi/Primary Examiner, Art Unit 3661
Prosecution Insights