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 .
Continued Examination Under 37 CFR 1.114
2. A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant’s submission filed on 07/10/2026 has been entered.
Examiner Notes
3. 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
Rejections under 35 U.S.C. § 101
4. In regard to the rejection of the claims under 35 U.S.C. § 101, Applicant's arguments filed 07/10/2026 have been fully considered and they are persuasive.
5. Applicant’s amendments have overcome the rejection of the claims under 35 U.S.C. § 101. Accordingly, the previous rejection of the claims under 35 USC § 101 is withdrawn.
Rejections under 35 U.S.C. § 103
6. In regard to the rejection of the claims under 35 U.S.C. § 103, Applicant's arguments filed 07/10/2026 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 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 independent claim(s) 9, and 17 has/have been amended similar to independent claim 1 and it/they is/are allowable for reasons similar to those presented in favor of patentability of claim 1.
10. This argument is unpersuasive as each independent claim has been fully rejected and for the reasons given above.
11. Applicant argues the dependent claim(s) is/are patentable by the virtue of its/their dependency on one of the independent claims and the additional features recited in the dependent claim(s).
12. 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 § 112
13. The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
14. Claim 1-6, 17-18, and 22-24 rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
15. Claim 1 recites the limitation “the largest curb” in line 16 of the claim. There is insufficient antecedent basis for this limitation in the claim. For the purpose of prior art rejection, the limitation was interpreted as “the curb”.
16. Claim(s) 2-6, 8, and 22-24 rejected by virtue of its/their dependency on claim 1.
17. Claim(s) 17 rejected for the same reason given above in regard to claim 1.
18. Claim(s) 18 rejected by virtue of its/their dependency on claim 17.
Claim Rejections - 35 USC § 103
19. 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.
20. Claim(s) 1, 6, 9, 14, 17, 19, and 23
is/are rejected under 35 U.S.C. 103 as being unpatentable over Silver et al. (US-9285230-B1) in view of Jiang et al. (US-20180225515-A1) and further in view of Yan et al. (US-20200317194-A1).
In regard to claim 1
, Silver discloses a method for detecting road features, comprising (Silver, in at least Fig. 3, Col 3, lines 4345, discloses a method designed to detect and locate road curbs and/or other barriers [i.e., detecting road features] in the vehicle's environment):
obtaining a point-cloud frame that comprises a description of an intensity of a reflection of beams from an area around an autonomous vehicle (Silver, in at least Fig. 3, Col 3, lines 27-28, Col 13, lines 6-12, Col 19, lines 59-60, discloses an autonomous or driverless vehicle [i.e., an autonomous vehicle], is configured to detect road curbs and/or other boundaries in the local environment of the vehicle during navigation. At block 302, the method 300 includes receiving a plurality of point clouds collected in an incremental order [i.e., obtaining a point-cloud frame] as a vehicle navigates a path where each point cloud includes data points that represent the environment of the vehicle at a given timepoint [i.e., an area around a vehicle] and has associated position information indicative of a position of the vehicle at the given timepoint. The computing system determines intensity information associated with laser returns captured by a vehicle sensor [i.e., a description of an intensity of a reflection of beams from an area around a vehicle]);
creating a plurality of clusters that each include (i) one or more seed points of the point-cloud frame, and (ii) additional points of the point-cloud frame added based on a relationship between the additional points and the one or more seed points, wherein the seeds points are selected for a region-growing process (Silver, in at least Fig. 3, Col. 3, lines 59-67, Col 14, lines 22-25, and lines 43-49, discloses after receiving one or multiple sets of data, the computing system processes the point clouds into an overall accumulated data format, a dense point cloud representation [i.e., cluster], based on respective associated position information of the point clouds [i.e., a relationship between the additional points and the one or more seed points]. The computing system fuses the point clouds into a dense point cloud representation that aligns the data points of the point clouds in an order that properly reflects the vehicle's environment by the incremental order [i.e., the seeds points are selected for a region-growing process]. At block 304, the method 300 includes, based on respective associated position information of the plurality of point clouds [i.e., one or more seed points of the point-cloud frame], processing, by a computing device, the plurality of point clouds into a dense point cloud representation [i.e., creating a plurality of clusters]. The dense point cloud representation enables the computing system to extract information that is not obtainable using a single point cloud of data, such as road curbs. The computing system formats the data into a dense point cloud representation to better reflect the position and/or orientation of objects in the environment of the vehicle [i.e., additional points of the point-cloud frame added]. Examiner notes, as mentioned above, the point clouds are processed into a dense point cloud representation based on respective associated position information of the point clouds. That is, the dense point cloud representation is formed based on the associated position information of the point clouds or proximity of the point clouds to other point clouds. Furthermore, the dense point cloud representation is generated by incremental order which means that a first point cloud (a random point cloud, which could be the first received point) is selected, and then a dense point cloud representation is created by adding the proximate point clouds to the dense point cloud based on their associated position information and proximity to the first point. Adding point to the dense cloud representation is the region-growing process and the selected first point is the seed point. Accordingly, Silver discloses creating a plurality of clusters that each include (i) one or more seed points of the point-cloud frame, and (ii) additional points of the point-cloud frame added based on a relationship between the additional points and the one or more seed points, wherein the seeds points are selected for a region-growing process);
detecting a road feature from the largest cluster (Silver, in at least Fig. 3, and Col 21, lines 44-47, discloses at block 310, the method 300 further includes, based on an output of the classification system, determining whether the given data points represent one or more road curbs [i.e., detecting a road feature from the largest cluster] in the environment of the vehicle); and
causing the autonomous vehicle to be driven to a destination according to detection results of the road feature (Silver, in at least Fig. 3, Col 21, lines 44-58, discloses at block 310, the method 300 further includes, based on an output of the classification system, determining whether the given data points represent one or more road curbs in the environment of the vehicle. The computing system determines any data (as originally captured and provided by vehicle sensors) that indicates the presence of road curbs or similar structures in the vehicle's environment based on outputs provided by a classifier or another computing entity. The vehicle's computing system use the outputs as provided by a classification system to locate road barriers (e.g., road curbs) in the vehicle's environment and navigate the vehicle based on the location of road barriers relative to the vehicle [i.e., causing the autonomous vehicle to be driven to a destination according to detection results of the road feature]. The vehicle navigates a path of travel based on avoiding a nearby road curb positioned on the side of a road);
wherein detecting the road feature comprises detecting, according to a boundary of the largest cluster, a curb of a road, wherein the autonomous vehicle is located on the road (Silver, in at least Fig. 4, Col 22, lines 39-45, discloses vehicle 400 uses a sensor or multiple sensors to detect segments of a road curb, such as road curb segment 404 of the road curb 402. Likewise, the vehicle 400 detects segments of multiple road curbs in some instances [i.e., wherein detecting the road feature comprises detecting, according to a boundary of the largest cluster, a curb of a road, wherein the vehicle is located on the road]. In addition, during operation, the vehicle's sensors detects other objects or boundaries in the environment as well, such as lane boundaries, guard-rails, etc. Examiner notes, as portrayed by Fig. 4, the vehicle detects the curb when the vehicle is located on the road);
While Silver, in at least Col 14, lines 50-54, discloses the computing system processes received point clouds into a dense point cloud representation by accumulating multiple point clouds and processing the point clouds into the representation, Silver does not explicitly disclose identifying from the plurality of clusters based on a total number of points included in each of the plurality of clusters, a largest cluster as a road surface;
wherein the curb of the road is determined by applying curve fitting to detected points corresponding to the largest curb;
wherein detected points corresponding to an obstruction that occludes the curb of the road are removed from consideration for curve fitting.
However, Jiang teaches identifying from the plurality of clusters based on a total number of points included in each of the plurality of clusters, a largest cluster as a road surface (Jiang, in at least [0062], teaches a mean grid map, a min grid map, and a max grid map corresponding to a dense point cloud under a world coordinate system are established. A threshold is selected to establish an undirected graph model with a slope between neighboring grids as a characteristic, to obtain two maximum blocks of communication areas as a candidate road surface [i.e., identifying from the plurality of clusters based on a total number of points included in each of the plurality of clusters, a largest cluster as a road surface]. Road surface grids adjacent to the candidate road surface are then queried. A threshold is selected to obtain a road surface point cloud in the road surface grid, and the road surface point cloud is filtered out, i.e., eliminating the road surface point cloud and the road edge point cloud in the dense point cloud. Examiner notes, the maximum blocks are the clusters and a maximum block necessarily includes a largest total number of point clouds);
wherein the curb of the road is determined by applying curve fitting to detected points corresponding to the largest curb (Jiang, in at least [0040], teaches after the dense point cloud and multiple frames of the sparse ordered point cloud are acquired, possible road edge points are obtained by processing the multiple frames of sparse ordered point cloud. The possible road edge points are subjected to three-dimensional spline curve fitting so as to construct a corresponding road edge model [i.e., the curb of the road is determined by applying curve fitting to detected points corresponding to the largest curb] corresponding to the laser point cloud based on the multiple frames of sparse ordered point cloud. Examiner notes, Spline fitting model is a curve fitting model);
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify silver in view of Jiang with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – curb detection – such that the maximum block with the largest number of point clouds is identified as the road surface and then the spline curve fitting is applied to determine the curb or the road edge and the combination would provide for improving the construction efficiency and accuracy of the road surface model (Jiang, see at least [0028]).
While Jiang, in at least [0048], teaches after repeating the operation above to obtain the road edge corner points corresponding to each frame of the sparse point cloud, the road edge corner points corresponding to all sparse point clouds are converted to a world coordinate system. After they are fused, noise is removed using a statistical filtering technique, the data amount is reduced using a point cloud dilution technique, and the road edge is repaired using the Karman filtering technique along the travelling track of the mobile vehicle. Afterwards, unordered corner points are fitted into a three-dimensional spline curve to obtain the road edge model corresponding to the laser cloud points, Silver, as modified by Jiang, does not explicitly teach wherein detected points corresponding to an obstruction that occludes the curb of the road are removed before the curve fitting.
However, Yan teaches wherein detected points corresponding to an obstruction that occludes the curb of the road are removed before the curve fitting (Yan, in at least Figs. 1-3, 6(c), and [0202], teaches the processing portion 19 subsequently identifies, in the 3D point cloud, the pixels corresponding to cluster CP in FIG. 6(c), as potential path pixels. The processing portion 19 attempts to identify a ‘ground plane’ of the path of the vehicle corresponding to the potential path pixels. The processing portion 19 then identifies edges of the path by eliminating potential path pixels at the lateral edges that are at relatively large heights above the ground plane. Thus, potential obstacles at the edges of the path such as bushes, trees, rocks or the like are eliminated [i.e., wherein detected points corresponding to an obstruction that occludes the curb of the road are removed]. The processing portion 19 then determines the location of lateral edges of the predicted path following elimination of lateral obstacles).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify silver, as already modified by Jian, in view of Yan with a reasonable expectation of success, as all inventions are directed to the same field of endeavor – curb detection – such that the potential obstacles at the edges of the path such as bushes, trees, rocks or the like are eliminated and then the three-dimensional spline curve fitting, as taught by Jiang, is applied to determine the curb of the road and the combination would provide for to improving the driving experience for the user by reducing workload (Yan, see at least [0004]).
In regard to claim 6
, Silver, as modified by Jiang and Yan, teaches the method of claim 1, wherein the road surface is detected from a portion of the area around the autonomous vehicle that is spanned by the largest cluster (Silver, in at least Col 9, lines 39-43, discloses the navigation and pathing system 148 is configured to determine a driving path for the vehicle 100 [i.e., wherein the road surface is detected from a portion of the area around the autonomous vehicle that is spanned by the largest cluster]. The navigation and pathing system 148 additionally is configured to update the driving path dynamically while the vehicle 100 is in operation).
In regard to claim 9
, Silver discloses an apparatus for detecting road features, comprising (Silver, in at least Col 1, line 30, discloses systems for detecting road curbs [i.e., an apparatus for detecting road features]):
a processor (Silver, in at least Fig. 1, and Col 5, lines 51-52, discloses the computing device 111 includes a processor 113 [i.e., a processor], and a memory 114); and
a memory comprising executable code that, when executed by the processor, causes the apparatus to (Silver, in at least Fig. 1, and Col 5, lines 51-55, discloses the computing device 111 includes a processor 113, and a memory 114 [i.e., a memory] which includes instructions 115 executable by the processor 113 [i.e., executable code that, when executed by the processor], and also stores map data 116):
obtain a point-cloud frame that comprises a description of an intensity of a reflection of beams from an area around an autonomous vehicle (Silver, in at least Fig. 3, Col 3, lines 27-28, Col 13, lines 6-12, Col 19, lines 59-60, discloses an autonomous or driverless vehicle [i.e., an autonomous vehicle], is configured to detect road curbs and/or other boundaries in the local environment of the vehicle during navigation. At block 302, the method 300 includes receiving a plurality of point clouds collected in an incremental order [i.e., obtain a point-cloud frame] as a vehicle navigates a path where each point cloud includes data points that represent the environment of the vehicle at a given timepoint [i.e., an area around a vehicle] and has associated position information indicative of a position of the vehicle at the given timepoint. The computing system determines intensity information associated with laser returns captured by a vehicle sensor [i.e., a description of an intensity of a reflection of beams from an area around a vehicle]);
create a plurality of clusters that each include (i) one or more seed points of the point-cloud frame, and (ii) additional points of the point-cloud frame added based on a relationship between the additional points and the one or more seed points, wherein the seeds points are selected for region-growing process (Silver, in at least Fig. 3, Col. 3, lines 59-67, Col 14, lines 22-25, and lines 43-49, discloses after receiving one or multiple sets of data, the computing system processes the point clouds into an overall accumulated data format, a dense point cloud representation [i.e., cluster], based on respective associated position information of the point clouds [i.e., a relationship between the additional points and the one or more seed points]. The computing system fuses the point clouds into a dense point cloud representation that aligns the data points of the point clouds in an order that properly reflects the vehicle's environment by the incremental order [i.e., the seeds points are selected for region-growing process]. At block 304, the method 300 includes, based on respective associated position information of the plurality of point clouds [i.e., one or more seed points of the point-cloud frame], processing, by a computing device, the plurality of point clouds into a dense point cloud representation [i.e., create a plurality of clusters]. The dense point cloud representation enables the computing system to extract information that is not obtainable using a single point cloud of data, such as road curbs. The computing system formats the data into a dense point cloud representation to better reflect the position and/or orientation of objects in the environment of the vehicle [i.e., additional points of the point-cloud frame added]. Examiner notes, as mentioned above, the point clouds are processed into a dense point cloud representation based on respective associated position information of the point clouds. That is, the dense point cloud representation is formed based on the associated position information of the point clouds or proximity of the point clouds to other point clouds. Furthermore, the dense point cloud representation is generated by incremental order which means that a first point cloud (a random point cloud, which could be the first received point) is selected, and then a dense point cloud representation is created by adding the proximate point clouds to the dense point cloud based on their associated position information and proximity to the first point. Adding point to the dense cloud representation is the region-growing process and the selected first point is the seed point. Accordingly, Silver discloses creating a plurality of clusters that each include (i) one or more seed points of the point-cloud frame, and (ii) additional points of the point-cloud frame added based on a relationship between the additional points and the one or more seed points, wherein the seeds points are selected for a region-growing process);
detect a road feature from the largest cluster (Silver, in at least Fig. 3, and Col 21, lines 44-47, discloses at block 310, the method 300 further includes, based on an output of the classification system, determining whether the given data points represent one or more road curbs [i.e., detecting a road feature from the cluster] in the environment of the vehicle); and
causing the autonomous vehicle to be driven to a destination according to detection results of the road feature (Silver, in at least Fig. 3, Col 21, lines 44-58, discloses at block 310, the method 300 further includes, based on an output of the classification system, determining whether the given data points represent one or more road curbs in the environment of the vehicle. The computing system determines any data (as originally captured and provided by vehicle sensors) that indicates the presence of road curbs or similar structures in the vehicle's environment based on outputs provided by a classifier or another computing entity. The vehicle's computing system use the outputs as provided by a classification system to locate road barriers (e.g., road curbs) in the vehicle's environment and navigate the vehicle based on the location of road barriers relative to the vehicle [i.e., causing the autonomous vehicle to be driven to a destination according to detection results of the road feature]. The vehicle navigates a path of travel based on avoiding a nearby road curb positioned on the side of a road);
wherein detecting the road feature comprises detecting, according to a boundary of the largest cluster, a curb of a road, wherein the autonomous vehicle is located on the road (Silver, in at least Fig. 4, Col 22, lines 39-45, discloses vehicle 400 uses a sensor or multiple sensors to detect segments of a road curb, such as road curb segment 404 of the road curb 402. Likewise, the vehicle 400 detects segments of multiple road curbs in some instances [i.e., wherein detecting the road feature comprises detecting, according to a boundary of the cluster, a curb of a road, wherein the vehicle is located on the road]. In addition, during operation, the vehicle's sensors detects other objects or boundaries in the environment as well, such as lane boundaries, guard-rails, etc. Examiner notes, as portrayed by Fig. 4, the vehicle detects the curb when the vehicle is located on the road);
While Silver, in at least Col 14, lines 50-54, discloses the computing system processes received point clouds into a dense point cloud representation by accumulating multiple point clouds and processing the point clouds into the representation, Silver does not explicitly disclose identify from the plurality of clusters based on a total number of points included in each of the plurality of clusters, a largest cluster as a road surface;
wherein the curb of the road is determined by applying curve fitting to detected points corresponding to the largest curb;
wherein detected points corresponding to an obstruction that occludes the curb of the road are removed before the curve fitting.
However, Jiang teaches identify from the plurality of clusters based on a total number of points included in each of the plurality of clusters, a largest cluster as a road surface (Jiang, in at least [0062], teaches a mean grid map, a min grid map, and a max grid map corresponding to a dense point cloud under a world coordinate system are established. A threshold is selected to establish an undirected graph model with a slope between neighboring grids as a characteristic, to obtain two maximum blocks of communication areas as a candidate road surface [i.e., identify from the plurality of clusters based on a total number of points included in each of the plurality of clusters, a largest cluster as a road surface]. Road surface grids adjacent to the candidate road surface are then queried. A threshold is selected to obtain a road surface point cloud in the road surface grid, and the road surface point cloud is filtered out, i.e., eliminating the road surface point cloud and the road edge point cloud in the dense point cloud. Examiner notes, the maximum blocks are the clusters and a maximum block necessarily includes a largest total number of point clouds);
wherein the curb of the road is determined by applying curve fitting to detected points corresponding to the largest curb (Jiang, in at least [0040], teaches after the dense point cloud and multiple frames of the sparse ordered point cloud are acquired, possible road edge points are obtained by processing the multiple frames of sparse ordered point cloud. The possible road edge points are subjected to three-dimensional spline curve fitting so as to construct a corresponding road edge model [i.e., the curb of the road is determined by applying curve fitting to detected points corresponding to the largest curb] corresponding to the laser point cloud based on the multiple frames of sparse ordered point cloud. Examiner notes, Spline fitting model is a curve fitting model);
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify silver in view of Jiang with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – curb detection – such that the maximum block with the largest number of point clouds is identified as the road surface and then the spline curve fitting is applied to determine the curb or the road edge and the combination would provide for improving the construction efficiency and accuracy of the road surface model (Jiang, see at least [0028]).
While Jiang, in at least [0048], teaches after repeating the operation above to obtain the road edge corner points corresponding to each frame of the sparse point cloud, the road edge corner points corresponding to all sparse point clouds are converted to a world coordinate system. After they are fused, noise is removed using a statistical filtering technique, the data amount is reduced using a point cloud dilution technique, and the road edge is repaired using the Karman filtering technique along the travelling track of the mobile vehicle. Afterwards, unordered corner points are fitted into a three-dimensional spline curve to obtain the road edge model corresponding to the laser cloud points, Silver, as modified by Jiang, does not explicitly teach wherein detected points corresponding to an obstruction that occludes the curb of the road are removed before the curve fitting.
However, Yan teaches wherein detected points corresponding to an obstruction that occludes the curb of the road are removed before the curve fitting (Yan, in at least Figs. 1-3, 6(c), and [0202], teaches the processing portion 19 subsequently identifies, in the 3D point cloud, the pixels corresponding to cluster CP in FIG. 6(c), as potential path pixels. The processing portion 19 attempts to identify a ‘ground plane’ of the path of the vehicle corresponding to the potential path pixels. The processing portion 19 then identifies edges of the path by eliminating potential path pixels at the lateral edges that are at relatively large heights above the ground plane. Thus, potential obstacles at the edges of the path such as bushes, trees, rocks or the like are eliminated [i.e., wherein detected points corresponding to an obstruction that occludes the curb of the road are removed]. The processing portion 19 then determines the location of lateral edges of the predicted path following elimination of lateral obstacles).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify silver, as already modified by Jian, in view of Yan with a reasonable expectation of success, as all inventions are directed to the same field of endeavor – curb detection – such that the potential obstacles at the edges of the path such as bushes, trees, rocks or the like are eliminated and then the three-dimensional spline curve fitting, as taught by Jiang, is applied to determine the curb of the road and the combination would provide for to improving the driving experience for the user by reducing workload (Yan, see at least [0004]).
In regard to claim 14
, Silver, as modified by Jiang and Yan, teaches the apparatus of claim 9.
Claim 14 recites an apparatus having substantially the same features of claim 6 above, therefore claim 14 is rejected for the same reasons as claim 6.
In regard to claim 17
, Silver discloses a non-transitory computer-readable medium storing a program that causes a computer to execute a process, the process comprising (Silver, in at least Col 1, lines 53-57, discloses a non-transitory computer readable medium having stored thereon executable instructions [i.e., a non-transitory computer-readable medium storing a program] that, upon execution by a computing device, cause the computing device to perform functions [i.e., that causes a computer to execute a process]):
obtaining a point-cloud frame that comprises a description of an intensity of a reflection of beams from an area around an autonomous vehicle (Silver, in at least Fig. 3, Col 3, lines 27-28, Col 13, lines 6-12, Col 19, lines 59-60, discloses an autonomous or driverless vehicle [i.e., an autonomous vehicle], is configured to detect road curbs and/or other boundaries in the local environment of the vehicle during navigation. At block 302, the method 300 includes receiving a plurality of point clouds collected in an incremental order [i.e., obtaining a point-cloud frame] as a vehicle navigates a path where each point cloud includes data points that represent the environment of the vehicle at a given timepoint [i.e., an area around a vehicle] and has associated position information indicative of a position of the vehicle at the given timepoint. The computing system determines intensity information associated with laser returns captured by a vehicle sensor [i.e., a description of an intensity of a reflection of beams from an area around a vehicle]);
creating a plurality of clusters that each include (i) one or more seed points of the point- cloud frame, and (ii) additional points of the point-cloud frame added based on a relationship between the additional points and the one or more seed points, wherein the seeds points are selected for region-growing process (Silver, Silver, in at least Fig. 3, Col. 3, lines 59-67, Col 14, lines 22-25, and lines 43-49, discloses after receiving one or multiple sets of data, the computing system processes the point clouds into an overall accumulated data format, a dense point cloud representation [i.e., cluster], based on respective associated position information of the point clouds [i.e., a relationship between the additional points and the one or more seed points]. The computing system fuses the point clouds into a dense point cloud representation that aligns the data points of the point clouds in an order that properly reflects the vehicle's environment by the incremental order [i.e., the seeds points are selected for region-growing process]. At block 304, the method 300 includes, based on respective associated position information of the plurality of point clouds [i.e., one or more seed points of the point-cloud frame], processing, by a computing device, the plurality of point clouds into a dense point cloud representation [i.e., creating a plurality of clusters]. The dense point cloud representation enables the computing system to extract information that is not obtainable using a single point cloud of data, such as road curbs. The computing system formats the data into a dense point cloud representation to better reflect the position and/or orientation of objects in the environment of the vehicle [i.e., additional points of the point-cloud frame added]. Examiner notes, as mentioned above, the point clouds are processed into a dense point cloud representation based on respective associated position information of the point clouds. That is, the dense point cloud representation is formed based on the associated position information of the point clouds or proximity of the point clouds to other point clouds. Furthermore, the dense point cloud representation is generated by incremental order which means that a first point cloud (a random point cloud, which could be the first received point) is selected, and then a dense point cloud representation is created by adding the proximate point clouds to the dense point cloud based on their associated position information and proximity to the first point. Adding point to the dense cloud representation is the region-growing process and the selected first point is the seed point. Accordingly, Silver discloses creating a plurality of clusters that each include (i) one or more seed points of the point-cloud frame, and (ii) additional points of the point-cloud frame added based on a relationship between the additional points and the one or more seed points, wherein the seeds points are selected for a region-growing process);
identifying from the plurality of clusters based on a total number of points included in each of the plurality of clusters, a largest cluster as a road surface; and
detecting a road feature from the largest cluster (Silver, in at least Fig. 3, and Col 21, lines 44-47, discloses at block 310, the method 300 further includes, based on an output of the classification system, determining whether the given data points represent one or more road curbs [i.e., detecting a road feature from the cluster] in the environment of the vehicle); and
causing the autonomous vehicle to be driven to a destination according to detection results of the road feature (Silver, in at least Fig. 3, Col 21, lines 44-58, discloses at block 310, the method 300 further includes, based on an output of the classification system, determining whether the given data points represent one or more road curbs in the environment of the vehicle. The computing system determines any data (as originally captured and provided by vehicle sensors) that indicates the presence of road curbs or similar structures in the vehicle's environment based on outputs provided by a classifier or another computing entity. The vehicle's computing system use the outputs as provided by a classification system to locate road barriers (e.g., road curbs) in the vehicle's environment and navigate the vehicle based on the location of road barriers relative to the vehicle [i.e., causing the autonomous vehicle to be driven to a destination according to detection results of the road feature]. The vehicle navigates a path of travel based on avoiding a nearby road curb positioned on the side of a road);
wherein detecting the road feature comprises detecting, according to a boundary of the largest cluster, a curb of a road, wherein the autonomous vehicle is located on the road (Silver, in at least Fig. 4, Col 22, lines 39-45, discloses vehicle 400 uses a sensor or multiple sensors to detect segments of a road curb, such as road curb segment 404 of the road curb 402. Likewise, the vehicle 400 detects segments of multiple road curbs in some instances [i.e., wherein detecting the road feature comprises detecting, according to a boundary of the cluster, a curb of a road, wherein the vehicle is located on the road]. In addition, during operation, the vehicle's sensors detects other objects or boundaries in the environment as well, such as lane boundaries, guard-rails, etc. Examiner notes, as portrayed by Fig. 4, the vehicle detects the curb when the vehicle is located on the road);
While Silver, in at least Col 14, lines 50-54, discloses the computing system processes received point clouds into a dense point cloud representation by accumulating multiple point clouds and processing the point clouds into the representation, Silver does not explicitly disclose identifying from the plurality of clusters based on a total number of points included in each of the plurality of clusters, a largest cluster as a road surface;
wherein the curb of the road is determined by applying curve fitting to detected points corresponding to the largest curb;
wherein detected points corresponding to an obstruction that occludes the curb of the road are removed from consideration for curve fitting.
However, Jiang teaches identifying from the plurality of clusters based on a total number of points included in each of the plurality of clusters, a largest cluster as a road surface (Jiang, in at least [0062], teaches a mean grid map, a min grid map, and a max grid map corresponding to a dense point cloud under a world coordinate system are established. A threshold is selected to establish an undirected graph model with a slope between neighboring grids as a characteristic, to obtain two maximum blocks of communication areas as a candidate road surface [i.e., identifying from the plurality of clusters based on a total number of points included in each of the plurality of clusters, a largest cluster as a road surface]. Road surface grids adjacent to the candidate road surface are then queried. A threshold is selected to obtain a road surface point cloud in the road surface grid, and the road surface point cloud is filtered out, i.e., eliminating the road surface point cloud and the road edge point cloud in the dense point cloud. Examiner notes, the maximum blocks are the clusters and a maximum block necessarily includes a largest total number of point clouds);
wherein the curb of the road is determined by applying curve fitting to detected points corresponding to the largest curb (Jiang, in at least [0040], teaches after the dense point cloud and multiple frames of the sparse ordered point cloud are acquired, possible road edge points are obtained by processing the multiple frames of sparse ordered point cloud. The possible road edge points are subjected to three-dimensional spline curve fitting so as to construct a corresponding road edge model [i.e., the curb of the road is determined by applying curve fitting to detected points corresponding to the largest curb] corresponding to the laser point cloud based on the multiple frames of sparse ordered point cloud. Examiner notes, Spline fitting model is a curve fitting model);
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify silver in view of Jiang with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – curb detection – such that the maximum block with the largest number of point clouds is identified as the road surface and then the spline curve fitting is applied to determine the curb or the road edge and the combination would provide for improving the construction efficiency and accuracy of the road surface model (Jiang, see at least [0028]).
While Jiang, in at least [0048], teaches after repeating the operation above to obtain the road edge corner points corresponding to each frame of the sparse point cloud, the road edge corner points corresponding to all sparse point clouds are converted to a world coordinate system. After they are fused, noise is removed using a statistical filtering technique, the data amount is reduced using a point cloud dilution technique, and the road edge is repaired using the Karman filtering technique along the travelling track of the mobile vehicle. Afterwards, unordered corner points are fitted into a three-dimensional spline curve to obtain the road edge model corresponding to the laser cloud points, Silver, as modified by Jiang, does not explicitly teach wherein detected points corresponding to an obstruction that occludes the curb of the road are removed before the curve fitting.
However, Yan teaches wherein detected points corresponding to an obstruction that occludes the curb of the road are removed before the curve fitting (Yan, in at least Figs. 1-3, 6(c), and [0202], teaches the processing portion 19 subsequently identifies, in the 3D point cloud, the pixels corresponding to cluster CP in FIG. 6(c), as potential path pixels. The processing portion 19 attempts to identify a ‘ground plane’ of the path of the vehicle corresponding to the potential path pixels. The processing portion 19 then identifies edges of the path by eliminating potential path pixels at the lateral edges that are at relatively large heights above the ground plane. Thus, potential obstacles at the edges of the path such as bushes, trees, rocks or the like are eliminated [i.e., wherein detected points corresponding to an obstruction that occludes the curb of the road are removed]. The processing portion 19 then determines the location of lateral edges of the predicted path following elimination of lateral obstacles).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify silver, as already modified by Jian, in view of Yan with a reasonable expectation of success, as all inventions are directed to the same field of endeavor – curb detection – such that the potential obstacles at the edges of the path such as bushes, trees, rocks or the like are eliminated and then the three-dimensional spline curve fitting, as taught by Jiang, is applied to determine the curb of the road and the combination would provide for to improving the driving experience for the user by reducing workload (Yan, see at least [0004]).
21. Claim(s) 2-3, 5, 10-11, 13 and 18
is/are rejected under 35 U.S.C. 103 as being unpatentable over Silver et al. (US-9285230-B1) in view of Jiang et al. (US-20180225515-A1) and further in view of Yan et al. (US-20200317194-A1) and further in view of Crouch et al. (US-20190370614-A1).
In regard to claim 2
, Silver, as modified by Jiang and Yan, teaches the method of claim 1, accordingly the rejection of claim 1 is incorporated.
Silver, as modified by Jiang and Yan, is silent on all claim limitations.
However, Crouch teaches further comprising:
determining that the additional points are related to the one or more seed points based on determining that the additional points are neighboring to the one or more seed points and meet a criterion associated with the one or more seed points (Crouch, in at least Fig. 6B, and [0067], teaches the 3D point cloud is obtained from a 3D scanner. Fig. 6B illustrates a segment 607 of the 3D point cloud 600 of Fig. 6A including a point 601 and nearest neighbor points 605 around the point 601 [i.e., determining that the additional points are related to the one or more seed points based on determining that the additional points are neighboring to the one or more seed points and meet a criterion associated with the one or more seed points]. Examiner notes, as mentioned above, neighbor points, or the additional points, are determined to be in the same cluster. In Fig. 6B, point 601 is the seed point).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Silver, as already modified by Jiang and Yan, in view of Crouch with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – classifying an object in a point cloud – and include the neighbor points in the same segment and the combination would provide for achieving acceptable range accuracy and detection sensitivity (Crouch, see at least [0004]).
In regard to claim 3
, Silver, as modified by Jiang and Yan and Crouch, teaches the method of claim 2, accordingly the rejection of claim 2 is incorporated.
Further, Crouch teaches wherein the additional points meet the criterion based on having values that are within a tolerance of values of the one or more seed points (Crouch, in at least Fig. 7, and [0097], teaches in step 717, the closest fit is performed between a test input point cloud and the model point clouds associated with the first and second classes, in order to determine whether or not the object should be classified in the first or second class. If the closest fit is too large, e.g., the mean square distance between points in the test input point cloud and points in the model point cloud for a minimum ratio of closest points between the point clouds is above a threshold square distance [i.e., wherein the additional points meet the criterion based on having values that are within a tolerance of values of the one or more seed points], then the object is considered not to belong to the class. If the mean square distance between points in the test input point cloud and points in the model point cloud for the top 90% of closest points between the point clouds is above 2 cm.sup.2, the object is considered not to belong to the class).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Silver, as already modified by Jiang and Yan and Crouch, further in view of Crouch with a reasonable expectation of success, as all inventions are directed to the same field of endeavor – classifying an object in a point cloud – and include the points that the mean square distance between points is below a threshold square distance and the combination would provide for achieving acceptable range accuracy and detection sensitivity (Crouch, see at least [0004]).
In regard to claim 5
, Silver, as modified Jiang and Yan, teaches the method of claim 1, accordingly the rejection of claim 1 is incorporated.
Silver, as modified Jiang and Yan, is silent on all limitations of the claim.
However, Crouch teaches the additional points included in a given cluster are determined using a k-dimensional tree constructed based on features of each point of the point-cloud frame (Crouch, in at least [0059 & 0012], discloses a k-d tree (short for k-dimensional tree) is a space-partitioning data structure for organizing points in a k-dimensional space. A k-d tree and NN (nearest neighbor) search is performed in order to assign class membership to any unknown object with a 3D point cloud [i.e., the additional points included in a given cluster are determined using a k-dimensional tree constructed based on features of each point of the point-cloud frame]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Silver, as already modified Jiang and Yan, in view of Crouch with a reasonable expectation of success, as all inventions are directed to the same field of endeavor – classifying an object in a point cloud – and use k-dimensional tree for assigning class membership and the combination would provide for achieving acceptable range accuracy and detection sensitivity (Crouch, see at least [0004]).
In regard to claim 10
, Silver, as modified by Jiang and Yan, discloses the apparatus of claim 9,
Claim 10 recites an apparatus having substantially the same features of claim 2 above, therefore claim 10 is rejected for the same reasons as claim 2.
In regard to claim 11
, Silver, as modified Jiang and Yan, teaches the apparatus of claim 9, accordingly the rejection of claim 9 is incorporated.
Silver, as modified by Jiang and Yan, is silent on all limitations of the claim.
However, Crouch teaches wherein a given cluster includes the additional points based on the additional points having feature values that are within a tolerance of corresponding feature values of the given cluster, wherein the feature values include a curvature value and a normal value (Crouch, in at least Figs. 6A-6B, 7, and [0068-0097], teaches in step 717, the closest fit is performed between a test input point cloud and the model point clouds associated with the first and second classes, in order to determine whether or not the object should be classified in the first or second class. If the closest fit is too large, e.g., the mean square distance between points in the test input point cloud and points in the model point cloud for a minimum ratio of closest points between the point clouds is above a threshold square distance [i.e., wherein a given cluster includes the additional points based on the additional points having feature values that are within a tolerance of corresponding feature values of the given cluster], then the object is considered not to belong to the class. If the mean square distance between points in the test input point cloud and points in the model point cloud for the top 90% of closest points between the point clouds is above 2 cm.sup.2, the object is considered not to belong to the class. Surface normals 602 [i.e., wherein the feature values include a curvature value and a normal value] are depicted in Fig. 6A and approximate a normal to the surface of the object at each point of the point cloud. Examiner notes, as depicted by Fig. 6A the surface normal are perpendicular to the curvature values. As such, the point clouds include a curvature value and a normal value).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Silver, as already modified by Jiang and Yan, in view of Crouch with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – classifying an object in a point cloud – and include the points that the mean square distance between points is below a threshold square distance and include the curvature value and a normal value in the point clouds and the combination would provide for achieving acceptable range accuracy and detection sensitivity (Crouch, see at least [0004]).
In regard to claim 13
, Silver, as modified by Jiang and Yan, teaches the apparatus of claim 9, accordingly the rejection of claim 9 is incorporated.
Silver, as modified by Jiang and Yan, is silent on all limitations of the claim.
However, Crouch teaches wherein construct a k-dimensional tree from the point-cloud frame (Crouch, in at least [0059], teaches a k-d tree (short for k-dimensional tree) is a space-partitioning data structure for organizing points in a k-dimensional space [i.e., wherein construct a k-dimensional tree from the point-cloud frame]), and
include the additional points in a given cluster based on a nearest neighbor search that uses the k-dimensional tree (Crouch, in at least Figs. 5A-5B, [0059 & 0062], teaches K-d trees are a useful data structure for several applications, such as searches involving a multidimensional search key (e.g. range searches and nearest neighbor searches). The k-d tree 550 is used to perform a nearest neighbor (NN) search, which aims to find the point in the set that is nearest to a given input point [i.e., include the additional points in a given cluster based on a nearest neighbor search that uses the k-dimensional tree]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Silver, as already modified by Jiang and Yan, in view of Crouch with a reasonable expectation of success, as all inventions are directed to the same field of endeavor – classifying an object in a point cloud – and include the neighbor points in the same segment and the combination would provide for achieving acceptable range accuracy and detection sensitivity (Crouch, see at least [0004]).
In regard to claim 18
, Silver, as modified by Jiang and Yan, teaches the non-transitory computer-readable medium of claim 17.
Claim 18 recites a non-transitory computer readable medium having substantially the same features of claim 2 above, therefore claim 18 is rejected for the same reasons as claim 2.
22. Claim(s) 4, and 12
is/are rejected under 35 U.S.C. 103 as being unpatentable over Silver et al. (US-9285230-B1) in view of Jiang et al. (US-20180225515-A1) and further in view of Yan et al. (US-20200317194-A1) and further in view of Liu et al. (CN-107123164-B).
In regard to claim 4
, Silver, as modified by Jiang and Yan, teaches the method of claim 1, wherein identifying the largest cluster includes:
identifying the largest cluster from remaining clusters of the plurality of clusters (Silver, in at least Col 14, lines 50-54, discloses the computing system processes received point clouds into a dense point cloud representation by accumulating multiple point clouds and processing the point clouds into the representation [i.e., identifying the largest cluster from remaining clusters of the plurality of clusters]).
Silver, as modified by Jiang and Yan, is silent on removing certain clusters of the plurality of clusters based on a growth of a certain clusters stopping during a region-growing process in which the additional points are added to each cluster, and
However, Liu teaches removing certain clusters of the plurality of clusters based on a growth of a certain clusters stopping during a region-growing process in which the additional points are added to each cluster (Liu, in at least [0076 & 0082] teaches the point cloud preprocessing module uses a smoothing and denoising method to remove noise from different point clouds, and performs clustering using an improved region growing method. Outliers in the point cloud are removed based on the clustering results. The point cloud preprocessing module corresponds to steps (1) and (2) in the three-dimensional reconstruction method. By removing outliers [i.e., removing certain clusters of the plurality of clusters] from the point cloud during the preprocessing stage using an improved region growing [i.e., based on a growth of a certain clusters stopping during a region-growing process in which the additional points are added to each cluster] method, not only can point cloud data that better matches the true shape of the original object be generated, but the impact of noise generated by outliers during subsequent point cloud registration can also be reduced).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Silver, as already modified by Jiang and Yan, in view of Liu with a reasonable expectation of success, as both inventions are directed to the same field of endeavor – classifying an object in a point cloud – and remove noise from different point clouds and perform clustering and the combination would provide for a three-dimensional reconstruction method and system that preserves sharp features (Liu, see at least [0002]).
In regard to claim 12
, Silver, as modified by Jiang and Yan, teaches the apparatus of claim 9.
Claim 12 recites an apparatus having substantially the same features of claim 4 above, therefore claim 12 is rejected for the same reasons as claim 4.
23. Claim(s) 8, and 16
is/are rejected under 35 U.S.C. 103 as being unpatentable over Silver et al. (US-9285230-B1) in view of Jiang et al. (US-20180225515-A1) and further in view of Yan et al. (US-20200317194-A1) and further in view of Sakai et al. (US-20180341019-A1).
In regard to claim 8
, Silver, as modified by Jiang and Yan, teaches the method of claim 1.
Silver, as modified by Jiang and Yan, is silent on all limitations of the claim.
However, Sakai teaches wherein obtaining the point-cloud frame comprises:
acquiring two consecutive single-frame point-clouds (Sakai, in at least Fig. 1, and [0061], teaches the point clustering module 127 compares the space, angle, respective distance, etc. between adjacent points [i.e., acquiring two consecutive single-frame point-clouds] in the point cloud to a threshold space, angle, distance, etc.); and
determine an accumulated point-cloud by registering the two consecutive single-frame point clouds into a common coordinate system, wherein the point-cloud frame includes the accumulated point-cloud (Sakai, in at least in at least Fig. 1, and [0061], teaches the point clustering module 127 compares the space, angle, respective distance, etc. between adjacent points [i.e., two consecutive single-frame point clouds] in the point cloud to a threshold space, angle, distance, etc. When the space, angle, distance, etc. exceeds (or is equal to) the threshold, the point clustering module 127 determines that the point(s) beyond the cluster 500b should be grouped [i.e., an accumulated point-cloud] with a separate cluster. Examiner notes, when the space, angle, respective distance, etc. between adjacent points is less than the threshold, then the adjacent points are clustered together which is determine an accumulated point-cloud by registering the two consecutive single-frame point clouds into a common coordinate system, wherein the point-cloud frame includes the accumulated point-cloud).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Silver, as already modified by Jiang and Yan, in view of Sakai with a reasonable expectation of success, as all inventions are directed to the same field of endeavor – classifying an object in a point cloud – and cluster the adjacent point clouds when the space, angle, respective distance, etc. between the adjacent points is less than a threshold and the combination would provide for forming 3D point clouds representing the vehicle's environment (Sakai, see at least [0002]).
In regard to claim 16
, Silver, as modified by Jiang and Yan, teaches the apparatus of claim 9.
Claim 16 recites an apparatus having substantially the same features of claim 8 above, therefore claim 16 is rejected for the same reasons as claim 8.
24. Claim(s) 22
is/are rejected under 35 U.S.C. 103 as being unpatentable over Silver et al. (US-9285230-B1) in view of Jiang et al. (US-20180225515-A1) and further in view of Yan et al. (US-20200317194-A1) Kee et al. (US-20180161986-A1).
In regard to claim 22
, Silver, as modified Jiang and Yan, teaches the method of claim 1, accordingly the rejection of claim 1 is incorporated.
Further, Jiang teaches wherein the method further comprises: smoothing the boundary of the largest cluster based on a spline function (Jiang, in at least [0040], teaches after the dense point cloud and multiple frames of the sparse ordered point cloud are acquired, possible road edge points are obtained by processing the multiple frames of sparse ordered point cloud. The possible road edge points are subjected to three-dimensional spline curve fitting so as to construct a corresponding road edge model [i.e., smoothing the boundary of the largest cluster based on a spline function] corresponding to the laser point cloud based on the multiple frames of sparse ordered point cloud. Examiner notes, Spline curve fitting necessarily smooths the boundary of a cluster that the curve fitting is applied to. As such, Jiang teaches smoothing the boundary of the largest cluster based on a spline function);
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Silver, as modified by Jiang and Yan, further in view of Jiang with a reasonable expectation of success, as all inventions are directed to the same field of endeavor – curb detection – and use a spline fitting model for smoothing the boundaries of the clusters, including the largest cluster, and the combination would provide for improving the construction efficiency and accuracy of the road surface model (Jiang, see at least [0028]).
Silver, as modified by Jiang and Yan, is silent on wherein detecting the curb of the road is further based on a concave hull of the largest cluster.
However, Kee teaches wherein detecting the curb of the road is further based on a concave hull of the largest cluster (Kee, in at least [0050], teaches once the factor graph is optimized, STORM projects the object models and background point clouds into a global coordinate frame. Point clouds of the background scene excludes object points to avoid aliasing with the object point clouds. These background point clouds are generated from the original sensor point clouds at each sensor pose. This is done by first computing the concave hull of the objects' database point clouds in the sensor frame. Then, all points inside the hull are removed from the background cloud [i.e., detecting the curb of the road is further based on a concave hull of the largest cluster, especially when the largest cluster is used]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Silver, as already modified by Jiang and Yan, in view of Kee with a reasonable expectation of success, as all inventions are directed to the same field of endeavor – identifying and localizing three-dimensional (3D) objects – and use concave hull to detect a three-dimensional object, such as curbs of a road and the combination would provide for improving localization and mapping (Kee, see at least [0004]).
25. Claim(s) 23
is/are rejected under 35 U.S.C. 103 as being unpatentable over Silver et al. (US-9285230-B1) in view of Jiang et al. (US-20180225515-A1) and further in view of Yan et al. (US-20200317194-A1) and further in view Non-patent Literature Hata et al. (“Robust Curb Detection and Vehicle Localization in Urban Environments”).
In regard to claim 23
, Silver, as modified by Jiang and Yan, teaches the method of claim 1, accordingly the rejection of claim 1 is incorporated.
Silver, as modified by Jiang and Yan, is silent on wherein the obstruction comprises another vehicle which is adjacent to the autonomous vehicle and is closer to the curb
However, Non-patent Literature Hata teaches wherein the obstruction comprises another vehicle which is adjacent to the autonomous vehicle and is closer to the curb (Non-patent Literature Hata, in at least p. 1259, teaches when obstacles as pedestrians and cars [i.e., wherein the obstruction comprises another vehicle which is adjacent to the autonomous vehicle and is closer to the curb, especially when the cars are parked beside a curb and the cars are adjacent to the vehicle] are present in the street, the distance filter will possibly detect them as curbs. These obstacles causes occlusion to the sensor and make difficult to identify actual curbs. The regression filter is introduced to estimate the curb shape and to remove points that do not follow the road model).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Silver, as already modified Jiang and Yan, in view of Non-patent Literature Hata with a reasonable expectation of success, as all inventions are directed to the same field of endeavor – curb detection – and detect other cars that are parked beside the curb and match the detected curb information against it to obtain accurate pose estimation, which is a fundamental capability to self-driving cars (Non-patent Literature Hata, see at least p. 1257).
26. Claim(s) 24
is/are rejected under 35 U.S.C. 103 as being unpatentable over Silver et al. (US-9285230-B1) in view of Jiang et al. (US-20180225515-A1) and further in view of Yan et al. (US-20200317194-A1) and further in view Sun et al. (US-20140254919-A1).
In regard to claim 24
, Silver, as modified by Jiang and Yan, teaches the method of claim 1, accordingly the rejection of claim 1 is incorporated.
Silver, as modified by Jiang and Yan, is silent on further comprising:
tracking a movement of an object in the road as the autonomous vehicle moves along the road; and
using tracking information to provide a priori information to the region-growing process.
However, Sun teaches further comprising:
tracking a movement of an object in the road as the autonomous vehicle moves along the road (Sun, in at least Fig. 1A, and [0046-0048], teaches in operation 110, an image processing device obtains a depth map of a successive 3D image over a period of time. In operation 120, the image processing device segments a moving object from the obtained depth map. In operation 130, the image processing device identifies and track the segmented moving object [i.e., tracking a movement of an object]); and
using tracking information to provide a priori information to the region-growing process (Sun, in at least Fig. 1B, and [0051], teaches in operation 121, an image processing device determines an initial seed point. In operation 122, the image processing device obtains a foreground object region. The image processing device performs region growing from the initial seed point to obtain the foreground object region [i.e., using tracking information to provide a priori information to the region-growing process]. In operation 124, the image processing device obtains a complete region of the moving object).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the application to modify Silver, as already modified Jiang and Yan, in view of Sun with a reasonable expectation of success, as all inventions are directed to the same field of endeavor – image processing – and use the method of Sun, to track the other moving objects on the road, and by determining an initial seed point and then perform region growing from the initial seed point, obtain a complete region of the moving object and the combination would provide for achieving a precise and stable image processing effect in a noisy environment (Jiang, see at least [0134]).
Conclusion
27. The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Mei et al. (US-20190163990-A1) teaches a system and method for large-scale lane marking detection using multimodal sensor data.
Poelman et al. (US-20170046589-A1) teaches pre-segmenting point cloud data to run real-time shape extraction.
Non-patent Literature Hata et al. (“Robust Curb Detection and Vehicle Localization in Urban Environments”) teaches curb detection by using a curve fitting model.
Ekin (US-20080199045-A1) teaches the selection of the road candidate pixels comprises identifying connected clusters of equal candidate pixel values in the mask, and selecting the largest coherent cluster.
28. 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.
29. 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.
30. 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.
31. 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, Ph.D./Primary Examiner, Art Unit 3661
Prosecution Insights