Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 3-5, 8-9, 11, 13-15, 18 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Pham (TRUNG PHAM ET AL: "NVAutoNet: Fast and Accurate 360^circ 3D Visual Perception For Self Driving", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 25 August 2023 (2023-08-25), XP091591250), in view of Qing (RAO QING ET AL: "In-Vehicle Object-Level 3D Reconstruction of Traffic Scenes", IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, IEEE, PISCATAWAY, NJ, USA, vol. 22, no. 12, 28 July 2020 (2020-07-28), pages 7747-7759, XP011889982, ISSN: 1524-9050, DOI: 10.1109/TITS.2020.3008080).
In re claim 1, Pham teaches A method comprising:
determining, by at least one processor, whether an ego has entered a park-eligible area (Section 4.3, page 7: “…the ability to localize and classify parking spaces.” and page 8: “The parking space detection task follows the same design (head, training strategy and losses) of the obstacle detection task. The parking detection network consists of classification and regression heads. The classification head predicts per-profile confidence scores. We rely on each parking spot's profile to implicitly encode its existence score. The regression head predicts the parking space oriented bounding boxes as discussed above.”);
reconstructing, by the processor, a space surrounding the ego by:
executing an artificial intelligence model using one or more cameras of the ego (SEE FIG. 1 depicting using images from cameras of ego vehicle fed into model) to predict a [distance value] for at least one occupied voxel within the ego's surrounding (Section 4.1, page 5, para [0001]: “3D object detection is a key capability for autonomous driving. The goal is to localize, classify and estimate dimensions and orientations of objects in 3D space. Each object is represented by its category and 3D cuboid…” and para [0002]: “…the 3D detection network will output… one object per grid cell.” and section 4.2, pages 6-7: “In a driving scenario, there is a lot more information which is relevant for safe driving beyond the predictions of 3D obstacle detection. For example, there can be random hazard obstacles like tyres, traffic cones lying on the road. Additionally, there are a lot of static obstacles like road-divider, road-side curb, and guard rails which are not covered by 3D obstacle detection… The region within the boundaries of the road which is not occupied by any obstacle could be carved out as a driveable region. The driveable region is used interchangeably as the freespace region. The down-stream behaviour planner would consume the freespace region information to plan a safe trajectory for the AV. So it is essential to have a perception component which predicts this freespace region. The freespace region is represented as a radial distance map (RDM).”);
identifying, by the processor, using the artificial intelligence model and the reconstructed space surrounding the ego, one or more parking spots within the park-eligible area (Section 4.3, page 7: “Another important aspect of autonomous driving is the ability to localize and classify parking spaces… Knowing the profile of every parking space is important for planning and control purposes. As such, every prediction output by our model will be assigned a parking profile. In the current system we support three different parking profiles: angled, parallel and perpendicular. As their name suggests, the profiles denote the types of planning and control maneuvering required to successfully park the car in the parking space.”);
receiving, by the at least one processor, a selection of at least one parking spot (Section 4.3, page 7: “Another important aspect of autonomous driving is the ability to localize and classify parking spaces… Knowing the profile of every parking space is important for planning and control purposes. As such, every prediction output by our model will be assigned a parking profile. In the current system we support three different parking profiles: angled, parallel and perpendicular. As their name suggests, the profiles denote the types of planning and control maneuvering required to successfully park the car in the parking space. Parallel parking spaces are ones that typically appear on the side of the street and require a parallel parking maneuver. Conversely, angled and perpendicular parking spots are ones where the car can be parked straight in (or backed in).”); and
transmitting, by the at least one processor, data associated with the selected parking spot to an autonomous navigation engine along with an instruction to navigate the ego and park the ego in the selected parking spot (Section 6, page 8: “NVAutoNet… optimizes for both accuracy and latency. Thirdly, NVAutoNet is tailored for real self-driving applications with a required detection range of up to 200 meters…” and pages 9-10: “Similar to object detection, we compute precision… strict criteria is necessary for real-world applications of autonomous parking as small misalignment between the detection and the actual parking space position can lead to imperfect parking.”; examiner notes the NVAutoNet system is “tailored for real self-driving applications” to implement a self-driving parking system given a parking spot, and is designed for “real-world applications of autonomous parking”. ).
Pham fails to teach a signed [distance value],
the signed distance value indicating a distance between the occupied voxel and a nearest occupied voxel.
However, Qing teaches a signed [distance value],
the signed distance value indicating a distance between the occupied voxel and a nearest occupied voxel (Section I.B, page 7748, para [0001]: “Our objective in this paper is to develop a resource-efficient solution to object-level 3D reconstruction, which is implementable on an automotive E/E architecture.” and para [0002]: “In addition, we focus on two specific use cases, namely in-vehicle augmented reality and automated parking.” and Section II.A, page 7748, para [0001]: “For representing shapes, there exist… implicit methods based on implicit functions such as the Signed Distance Function (SDF) [15], [16]. The major advantage of SDF over explicit shape description or other implicit methods is that the normal vectors of the object surface can be easily calculated for further use, e.g., shading. Also, a fast approach exists to render the 3D geometry from an SDF [17], which is suitable for in-vehicle use cases. Therefore, we narrowed our focus to 3D reconstruction methods using SDFs for the literature review related to this work.”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Pham to incorporate the teachings of Qing to provide a signed [distance value], the signed distance value indicating a distance between the occupied voxel and a nearest occupied voxel with the NVAutoNet: Fast and Accurate 360^circ 3D Visual Perception For Self Driving of Pham. Doing so enables a resource-efficient solution to object-level 3D reconstruction, and that the normal vectors of the object surface can be easily calculated for further use, e.g., shading, as recognized by Qing (Section I.B, page 7748, para [0001] and Section II.A, page 7748, para [0001]).
System claim 11 and system claim 18 are rejected for the same reasons as method claim 1 for having similar limitations and being similar in scope.
In re claim 3, Pham and Qing teach all of the limitations of claim 1 stated above where Pham further teaches wherein the at least one processor determines whether the ego has entered the park-eligible area using a second artificial intelligence model that ingest data received from the one or more cameras of the ego (Section 4.3, page 7, para [0001]: “Another important aspect of autonomous driving is the ability to localize and classify parking spaces. Each parking space is represented as an oriented rectangle… Knowing the profile of every parking space is important for planning and control purposes. As such, every prediction output by our model will be assigned a parking profile. In the current system we support three different parking profiles: angled, parallel and perpendicular. As their name suggests, the profiles denote the types of planning and control maneuvering required to successfully park the car in the parking space. Parallel parking spaces are ones that typically appear on the side of the street and require a parallel parking maneuver. Conversely, angled and perpendicular parking spots are ones where the car can be parked straight in (or backed in).” and page 8, para [0002]: “The parking space detection task follows the same design (head, training strategy and losses) of the obstacle detection task. The parking detection network consists of classification and regression heads. The classification head predicts per-profile confidence scores. We rely on each parking spot's profile to implicitly encode its existence score. The regression head predicts the parking space oriented bounding boxes as discussed above.”).
System claim 13 and system claim 20 are rejected for the same reasons as method claim 3 for having similar limitations and being similar in scope.
In re claim 4, Pham and Qing teach all of the limitations of claim 3 stated above where Pham further teaches wherein the second artificial intelligence model determines whether the ego has entered the park-eligible area based on an orientation of other vehicles within the park- eligible area (Section 6.4.1, page 10, paras [0001]-[0002]: “Presented in Table 5, the obstacle detection accuracy outcomes are showcased… The network demonstrates reasonable accuracy in estimating orientations for vehicles and trucks, displaying average errors below 7 degrees… For more detailed insights into detection results, Table 6 provides a granular breakdown. Generally, detection accuracy drops steadily as distances increase.”).
System claim 14 is rejected for the same reasons as method claim 4 for having similar limitations and being similar in scope.
In re claim 5, Pham and Qing teach all of the limitations of claim 1 stated above where Pham further teaches wherein the one or more parking spots are selected based on a respective path attribute from the ego to the one or more parking spots (Section 4.3, page 7, para [0001]: “Knowing the profile of every parking space is important for planning and control purposes… As their name suggests, the profiles denote the types of planning and control maneuvering required to successfully park the car in the parking space. Parallel parking spaces are ones that typically appear on the side of the street and require a parallel parking maneuver. Conversely, angled and perpendicular parking spots are ones where the car can be parked straight in (or backed in).”).
System claim 15 is rejected for the same reasons as method claim 5 for having similar limitations and being similar in scope.
In re claim 8, Pham and Qing teach all of the limitations of claim 1 stated above where Qing further teaches further comprising revising, by the at least one processor, a visual attribute of the selected parking spot (Section II.A, page 7750, para [0003]: “In practice, an SDF is discretized and stored as a floating- point array. Each element of the array records the distance from its grid location to the zero-level. The higher the dimen- sion of the array is, the finer is the discretization of the original geometry. Fig. 5 illustrates how the fineness of an SDF is affected by the dimension of the array container. As can be observed in Fig. 5, more delicate details in the concave parts of the original 3D model (mirrors, wheels) are not captured in the SDFs.”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Pham and Qing to further incorporate the teachings of Qing to provide further comprising revising, by the at least one processor, a visual attribute of the selected parking spot with the NVAutoNet: Fast and Accurate 360^circ 3D Visual Perception For Self Driving of Pham as modified by Qing. Doing so enables finer the discretization of the original geometry, as recognized by Qing (Section II.A, page 7750, para [0003]).
In re claim 9, Pham and Qing teach all of the limitations of claim 1 stated above where Pham further teaches wherein at least one parking spot requires parallel parking the ego (Section 4.3, page 7, para [0001]: “Knowing the profile of every parking space is important for planning and control purposes… As their name suggests, the profiles denote the types of planning and control maneuvering required to successfully park the car in the parking space. Parallel parking spaces are ones that typically appear on the side of the street and require a parallel parking maneuver. Conversely, angled and perpendicular parking spots are ones where the car can be parked straight in (or backed in).”).
Claims 2, 6, 10, 12, 16 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Pham (TRUNG PHAM ET AL: "NVAutoNet: Fast and Accurate 360^circ 3D Visual Perception For Self Driving", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 25 August 2023 (2023-08-25), XP091591250), in view of Qing (RAO QING ET AL: "In-Vehicle Object-Level 3D Reconstruction of Traffic Scenes", IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, IEEE, PISCATAWAY, NJ, USA, vol. 22, no. 12, 28 July 2020 (2020-07-28), pages 7747-7759, XP011889982, ISSN: 1524-9050, DOI: 10.1109/TITS.2020.3008080) and further in view of Minster (US Patent No. 20170267233 A1).
In re claim 2, Pham and Qing teach all of the limitations of claim 1 stated above but fails to teach wherein the at least one processor determines whether the ego has entered the park-eligible area based on at least one of a location of the ego matching a park-eligible location, identifying a sign within the space surrounding the ego indicating the park-eligible area, or a speed of the ego.
However, Minster teaches wherein the at least one processor determines whether the ego has entered the park-eligible area based on at least one of a location of the ego matching a park-eligible location, identifying a sign within the space surrounding the ego indicating the park-eligible area, or a speed of the ego (Paras [0062]-[0063]: “S132 preferably includes processing parking space data to identify parking spaces using visual indicators of parking spaces. Alternatively, S132 may include processing parking space data in any manner.” “For example, S132 may include analyzing image data captured by cameras of an autonomous vehicle to identify occupied parking spaces (e.g., by detecting non-moving vehicles on sides of the road), temporarily occupied parking spaces (e.g., by detecting a person or a dog blocking a parking spot), and unoccupied parking spaces (e.g., by detecting parking meters, by detecting road markers, by detecting unoccupied sections of curb, etc.). As another example, S132 may include analyzing image data to identify and interpret parking signs.”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Pham and Qing to further incorporate the teachings of Minster to provide wherein the at least one processor determines whether the ego has entered the park-eligible area based on at least one of a location of the ego matching a park-eligible location, identifying a sign within the space surrounding the ego indicating the park-eligible area, or a speed of the ego with the NVAutoNet: Fast and Accurate 360^circ 3D Visual Perception For Self Driving of Pham as modified by Qing. Doing so increases the ability of the autonomous vehicles to be utilized efficiently during peak and off-peak times, meet ridership demands, satisfy predetermined maintenance or similar schedules, and the like, as recognized by Minster (Para [0014]).
System claim 12 and system claim 19 are rejected for the same reasons as method claim 2 for having similar limitations and being similar in scope.
In re claim 6, Pham and Qing teach all of the limitations of claim 1 stated above but fails to teach wherein the one or more parking spots are selected based on paint line associated with each parking spot.
However, Minster teaches wherein the one or more parking spots are selected based on paint line associated with each parking spot (Para [0064]: “S132 preferably performs image analysis using model-based feature detection (e.g., comparing image data to examples of features known to correspond to parking space characteristics and/or status), but may additionally or alternatively include performing feature detection in any manner (e.g., via machine learning algorithms). Features detected by S132 that may correspond to parking space characteristics and/or status may include… the presence of painted parking space demarcations on the street…”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Pham and Qing to further incorporate the teachings of Minster to provide wherein the one or more parking spots are selected based on paint line associated with each parking spot with the NVAutoNet: Fast and Accurate 360^circ 3D Visual Perception For Self Driving of Pham as modified by Qing. Doing so increases the ability of the autonomous vehicles to be utilized efficiently during peak and off-peak times, meet ridership demands, satisfy predetermined maintenance or similar schedules, and the like, as recognized by Minster (Para [0014]).
System claim 16 is rejected for the same reasons as method claim 6 for having similar limitations and being similar in scope.
In re claim 10, Pham and Qing teach all of the limitations of claim 1 stated above but fails to teach further comprising:
displaying, by the at least one processor, a visual indicator for at least one identified parking spot.
However, Minster teaches further comprising:
displaying, by the at least one processor, a visual indicator for at least one identified parking spot (Para [0053]: “In use, the raw parking space data maybe used by the autonomous vehicle and/or associated autonomous vehicle platform to generate useful parking space information. For example, the autonomous vehicle can use the raw parking data of multiple available parking spaces to generate a heat map of parking spaces in a geographic area surrounding the autonomous vehicle. Such a heat map may display available and unavailable parking spaces, as well as other related parking space data. Additionally, or alternatively, the raw parking space data can be converted into schematics or one or more schematic drawings by the autonomous vehicle or an associated platform. The schematic drawing can be a top-down view of a parking space and a surrounding of the autonomous vehicle together with a rendering of the position of the autonomous vehicle relative to the parking space. This example parking schematic may be used by a passenger or even a remote expert attempting to assist the autonomous vehicle with parking into a parking space.”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Pham and Qing to further incorporate the teachings of Minster to provide further comprising: displaying, by the at least one processor, a visual indicator for at least one identified parking spot with the NVAutoNet: Fast and Accurate 360^circ 3D Visual Perception For Self Driving of Pham as modified by Qing. Doing so generates useful parking space information used by a passenger or even a remote expert attempting to assist the autonomous vehicle with parking into a parking space, as recognized by Minster (Para [0053]).
Claims 7 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Pham (TRUNG PHAM ET AL: "NVAutoNet: Fast and Accurate 360^circ 3D Visual Perception For Self Driving", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 25 August 2023 (2023-08-25), XP091591250), in view of Qing (RAO QING ET AL: "In-Vehicle Object-Level 3D Reconstruction of Traffic Scenes", IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, IEEE, PISCATAWAY, NJ, USA, vol. 22, no. 12, 28 July 2020 (2020-07-28), pages 7747-7759, XP011889982, ISSN: 1524-9050, DOI: 10.1109/TITS.2020.3008080) and further in view of Park (US Patent No. 20220292971 A1).
In re claim 7, Pham and Qing teach all of the limitations of claim 1 stated above but fails to teach wherein the one or more parking spots are selected based on whether each parking spot includes a shaped group of painted voxels within its driving surface.
However, Park teaches wherein the one or more parking spots are selected based on whether each parking spot includes a shaped group of painted voxels within its driving surface (Para [0122]: “It should be understood that various embodiments of the specification and terms used therefor are not intended to limit the technology described in the specification to specific embodiments, but include various changes, equivalents and/or substitutions of the embodiments.”, para [0382]: “For example, the processor 1608 may identify a parking slot located in a parking lot from a parking lot image obtained through an image obtaining sensor, and also identify whether or not a vehicle is parked in the parking slot. For example, when a parking line marked in the parking lot is detected through analysis of the parking lot image obtained through the image obtaining sensor, the processor 1608 may identify a detected area as a parking slot, and determine whether or not parking is possible according to whether or not a vehicle exists in the identified parking slot.” and paras [0533]-[0534]: “As another example, when the depth estimation unit 3704 estimates a depth of the selected pixel value, voxelization of the input image may be used. The method of estimating a depth of the pixel value using voxelization of the image by the depth estimation unit 3704 will be described with reference to the following FIG. 39.” “A voxel indicates a value of a regular grid in a 3D space in a medical and science field and this values are used as very important elements to analyze and visualize the data. This is because the voxel is an array of a volume element which configures a notational 3D space and is generally used in a computer based modeling and graphic simulation. In the 3D printing, the voxel is widely used because of its own depth. In one embodiment, the voxelization divides point cloud into 3D voxels having the same space and converts points in a predetermined group in each voxel into a unified feature representation through a voxel feature encoding (VFE) layer.”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Pham and Qing to further incorporate the teachings of Park to provide wherein the one or more parking spots are selected based on whether each parking spot includes a shaped group of painted voxels within its driving surface with the NVAutoNet: Fast and Accurate 360^circ 3D Visual Perception For Self Driving of Pham as modified by Qing. Doing so enables converting points in a predetermined group in each voxel into a unified feature representation through a voxel feature encoding (VFE) layer, as recognized by Park (Para [0534]).
System claim 17 is rejected for the same reasons as method claim 7 for having similar limitations and being similar in scope.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JAMES EDWARD MUNION whose telephone number is (571)270-0437. The examiner can normally be reached Monday-Friday 7:30-5:00.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Steven Lim can be reached at 571-270-1210. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/JAMES E MUNION/Examiner, Art Unit 2688 06/11/2026
Prosecution Insights