DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
Response to Amendment
This FINAL action is in response to amendment filed on 10/08/2025.
Claims 1, 4-9, 12, 16-18 are amended.
Claims 2-3, 10-11, 13-15 are original.
Claims 19-35 are withdrawn.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-6, 9-10 is/are rejected under 35 U.S.C. 103 as being unpatentable over US 20200410702 A1 Zhang; Ronghua et al. (hereinafter Zhang), in view of US 20230027369 A1 Schmitt; Paul et al. (hereinafter Schmitt), further in view of KR 20220137239 A PARK MIN SIK et al. (hereinafter Park), and further in view of US 20140002597 A1 Taguchi; Yuichi et al. (hereinafter Taguchi).
Regarding claim 1, Zhang discloses:
A method for navigating an autonomous vehicle (see Zhang at least [0078] the techniques disclosed herein may help improve the self-driving of vehicles) comprising:
organizing a first map associated with a current location of the autonomous vehicle, forming current organized data (see Zhang at least [0082] FIGS. 6A-6B illustrate example geographical regions 610a and 610b that may be defined in an HD map according to one or more embodiments... The online HD map system 110 may store data in a representation of a geographical region that may allow for transitions from one geographical region to another as a vehicle 150 drives across geographical region boundaries and Fig. 6B geographical region 610a);
organizing a second map associated with a potential location to which the autonomous vehicle could navigate, forming potential location organized data (see Zhang at least [0082] FIGS. 6A-6B illustrate example geographical regions 610a and 610b that may be defined in an HD map according to one or more embodiments... The online HD map system 110 may store data in a representation of a geographical region that may allow for transitions from one geographical region to another as a vehicle 150 drives across geographical region boundaries and [0084] Accordingly, the vehicle computing system 120 of the corresponding vehicle 150 may continue to use the geographical region 610a as the current geographical region of the vehicle. Once the corresponding vehicle 150 crosses the boundary 620 of the buffer at location 650c, the vehicle computing system 120 may switch the current geographical region of the corresponding vehicle 150 to geographical region 610b from geographical region 610a and Fig. 6B geographical region 610b);
updating the current organized data as the autonomous vehicle navigates based on the potential location organized data and the current location (see Zhang at least [0039] The vehicles 150 may be configured to provide the sensor data 115 that may be captured while driving along various routes and to send it to the online HD map system 110. The online HD map system 110 may be configured to use the sensor data 115 received from the vehicles 150 to create and update HD maps describing the regions in which the vehicles 150 may be driving and [0038] the sensor data may include information that may describe the current state of the vehicle 150, the location and motion parameters of the vehicles 150, etc.);
selectively updating the potential location organized data based on a movement direction, a movement speed, the potential location organized data, and the current location (see Zhang at least [0039] The online HD map system 110 may be configured to use the sensor data 115 received from the vehicles 150 to create and update HD maps describing the regions in which the vehicles 150 may be driving. The online HD map system 110 may be configured to build high definition maps based on the collective sensor data 115 that may be received from the vehicles 150 and to store the HD map information in the HD map store 165 and [0038] the sensor data may include information that may describe the current state of the vehicle 150, the location and motion parameters of the vehicles 150, etc. and [0046] motion data of the vehicle 150 such as velocity, acceleration, direction of movement, speed, angular rate, and so on);
filtering real-time data received by the autonomous vehicle, forming real-time data (see Zhang at least [0100] one or more of the images may be closer to each other than a determined minimum distance and the system 900 may filter and/or remove the images from selection as keyframes such that the distance between the keyframes may be greater than the threshold distance); and
navigating the vehicle based on the localized pose (see Zhang at least [0020] Embodiments of the present disclosure may perform global visual localization of an autonomous vehicle using keyframes and keypoints and may create, maintain, and store high definition (HD) maps that may include up-to-date information with high accuracy or precision given an initial estimation of a pose and one or more images captured by one or more cameras of an autonomous vehicle. The HD maps may be used by an autonomous vehicle to safely navigate to various destinations without human input or with limited human input).
Zhang does not teach: scan matching the processed current organized data and the real-time data, forming matched map points based on temporally and positionally dynamic thresholds; rejecting outlying data from the matched map points based on feature properties, the dynamic thresholds, and an outlier determination algorithm, forming a pose estimation; and correcting the pose estimation based on planar features associated with the current location, forming a localized pose.
However, Schmitt teaches: scan matching the processed current organized data and the real-time data, forming matched map points based on temporally and positionally dynamic thresholds (See Schmitt at least [0021] the threshold may be increased when location has not been set for a longer period of time and [0024] the set point update criteria match the criteria for accurately determining location (e.g., that n of m pipelines establish distances within a threshold distance of one another) and [0026] While examples are provided herein with reference to specific localization pipelines (e.g., LiDAR-based, radar-based, dead reckoning-based, etc.), embodiments of the present disclosure may use any combination of various localization pipelines… global localization, which attempts to determine a location of a vehicle within a known space (e.g., a map of a given area)).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the image-based vehicle localization method disclosed by Zhang to include the dynamic thresholds of Schmitt. One of ordinary skill in the art would have been motivated to make this modification because thresholds designed for different types of situations can overcome faults of some localization means, as suggested by Schmitt (see Schmitt at least [0022] use of a dynamic threshold period (based, e.g., on the factors noted above) can account for known issues with localization pipelines. For example, the system may continue to function even when a LiDAR-based pipeline experiences a temporary “blip” in operation).
Zhang and Schmitt do not teach: rejecting outlying data from the matched map points based on feature properties, the dynamic thresholds, and an outlier determination algorithm, forming a pose estimation; and correcting the pose estimation based on planar features associated with the current location, forming a localized pose.
However, Park teaches: rejecting outlying data from the matched map points based on feature properties, the dynamic thresholds, and an outlier determination algorithm, forming a pose estimation (see Park at least [0032] The location information refinement unit (108) can remove outliers and collect inliers based on a threshold value for feature matching pairs between initialized reference and target camera images. 378 In addition, the location information refining unit (108) can change the threshold value and collect refined inliers based on the changed threshold value for the collected feature matching pairs at least once until the location information estimate value of the feature matching pairs collected exceeds the set error, and is determined to be less than the set error).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the image-based vehicle localization method disclosed by Zhang and Schmitt to include the outlier rejection step of Park. One of ordinary skill in the art would have been motivated to make this modification because accurately estimating a pose requires the elimination of outlying data, which may otherwise lead to faulty outcomes, as suggested by Park (see Park at least [0003] the performance of location information estimation depends on the effective removal of outliers).
Zhang, Schmitt, and Park do not teach: correcting the pose estimation based on planar features associated with the current location, forming a localized pose.
However, Taguchi teaches: correcting the pose estimation based on planar features associated with the current location, forming a localized pose (see Taguchi at least [0023] Given the map, we use a prediction-and-correction framework to estimate the pose of the current frame: We predict the pose of the camera, and use the pose to determine correspondences between the point and plane measurements and the point and plane landmarks, which are then used to determine the camera pose).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the image-based vehicle localization method disclosed by Zhang, Schmitt, and Park to include the point-to-plane based pose estimate correction technique of Taguchi. One of ordinary skill in the art would have been motivated to make this modification because using the correspondence of points and nearby planes allows a vehicle to continuously update and improve its pose estimation for the best results, as suggested by Taguchi (see Taguchi at least [0014] The method incorporates relocalization and bundle adjustment processes using both the points and planes to recover from tracking failures and to continuously refine camera pose estimates).
Regarding claim 2, Zhang, Schmitt, Park, and Taguchi teach: The method as in claim 1 further comprising receiving the first map from a source remote to the autonomous vehicle (see Zhang at least [0043] In some embodiments, the online HD map system 110 may determine that the particular vehicle 150 may not have certain portions of the HD map data stored locally in a local HD map store of the particular vehicle computing system 120 of the particular vehicle 150. In these or other embodiments, in response to such a determination, the online HD map system 110 may be configured to send a particular portion of the HD map data to the vehicle 150 and [0082] geographical regions 610a and 610b that may be defined in an HD map).
Regarding claim 3, Zhang, Schmitt, Park, and Taguchi teach: The method as in claim 1 further comprising accessing the first map from a database stored local to the autonomous vehicle (see Zhang at least [0058] The corresponding vehicle 150 may store the HD map data in the local HD map store 275. The modules of the vehicle computing system 120 may interact with the map data using an HD map API 205 and [0082] geographical regions 610a and 610b that may be defined in an HD map).
Regarding claim 4, Zhang, Schmitt, Park, and Taguchi teach: The method as in claim 1 further comprising receiving the second map from a source remote to the autonomous vehicle (see Zhang at least [0043] In some embodiments, the online HD map system 110 may determine that the particular vehicle 150 may not have certain portions of the HD map data stored locally in a local HD map store of the particular vehicle computing system 120 of the particular vehicle 150. In these or other embodiments, in response to such a determination, the online HD map system 110 may be configured to send a particular portion of the HD map data to the vehicle 150 and [0082] geographical regions 610a and 610b that may be defined in an HD map).
Regarding claim 5, Zhang, Schmitt, Park, and Taguchi teach: The method as in claim 1 further comprising accessing the second map from a database stored local to the autonomous vehicle (see Zhang at least [0058] The corresponding vehicle 150 may store the HD map data in the local HD map store 275. The modules of the vehicle computing system 120 may interact with the map data using an HD map API 205 and [0082] geographical regions 610a and 610b that may be defined in an HD map).
Regarding claim 6, Zhang, Schmitt, Park and Taguchi teach: The method as in claim 1 wherein said selectively updating comprises: determining if the autonomous vehicle is located in a border region between the first map and the second map (see Zhang at least [0084] The corresponding vehicle 150 may traverse along a route to reach a location 650b where it may cross the boundary of the geographical region 610 but may stay within the boundary 620 of the buffer and Fig. 6B); and
updating the potential location organized data when the autonomous vehicle navigates outside of the border region and the first map (see Zhang at least [0084] Once the corresponding vehicle 150 crosses the boundary 620 of the buffer at location 650c, the vehicle computing system 120 may switch the current geographical region of the corresponding vehicle 150 to geographical region 610b from geographical region 610a and Fig. 6B).
Regarding claim 9, Zhang, Schmitt, Park, and Taguchi teach: The method as in claim 1 further comprising segregating the first map and the second map into planar points and non-planar points (see Taguchi at least [0047] the point and plane measurements located as inliers in the RANSAC-based registration are associated to corresponding landmarks, while those located as outliers are discarded…The additional plane measurements are extracted by using a RANSAC-based plane fitting on pixels that are not inliers of any existing plane measurements. The additional point and plane measurements are added as new landmarks to the map).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the image-based vehicle localization method disclosed by Zhang, Schmitt, Park, and Taguchi to include the plane correspondence determination process of Taguchi. One of ordinary skill in the art would have been motivated to make this modification because determining the presence and correspondence of planes between datasets reduces the amount of noise in the localization process, as suggested by Taguchi (see Taguchi at least [0045] The procedure prioritizes plane correspondences over point correspondences, because the number of planes is typically much smaller than the number of points, and planes have less noise due to the support from many points).
Regarding claim 10¸ Zhang, Schmitt, Park, and Taguchi teach: The method as in claim 9 further comprising removing the non-planar points (see Taguchi at least [0047] the point and plane measurements located as inliers in the RANSAC-based registration are associated to corresponding landmarks, while those located as outliers are discarded).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the image-based vehicle localization method disclosed by Zhang, Schmitt, Park, and Taguchi to include the plane correspondence determination and non-planar point elimination process of Taguchi. One of ordinary skill in the art would have been motivated to make this modification because removing non-planar points reduces the amount of noise in the localization process by focusing on planes, as suggested by Taguchi (see Taguchi at least [0045] The procedure prioritizes plane correspondences over point correspondences, because the number of planes is typically much smaller than the number of points, and planes have less noise due to the support from many points).
Claim(s) 7 and 8 is/are rejected under 35 U.S.C. 103 as being unpatentable over Zhang, in view of Schmitt, further in view of Park, further in view of Taguchi, and further in view of WO 2020009826 A1 HUBER STEVEN (hereinafter Huber).
Regarding claim 7, Zhang, Schmitt, Park, and Taguchi teach: The method as in claim 1.
Zhang, Schmitt, Park, and Taguchi do not teach: wherein said filtering comprises downsampling the real-time data according to pre-selected criteria, forming downsampled data.
Huber teaches wherein said filtering comprises downsampling the real-time data according to pre-selected criteria, forming downsampled data (see Huber at least [0024] Down-sampling the point cloud data spatially, such as by minimum distance between points to even out density, sensor noise, and fine features that are not important in large plane detection).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the vehicle localization method disclosed by Zhang, Schmitt, Park, and Taguchi to include the data down-sampling of Huber. One of ordinary skill in the art would have been motivated to make this modification because down-sampling the data that is used to identify planes from point cloud data allows the best data to be considered and create the most accurate planes, as suggested by Huber (see Huber at least [0023] Increased likelihood of a best possible solution can be achieved by either increasing the number of iterations or, more interestingly, reducing the number of points that are considered by RANSAC to the most likely candidates).
Regarding claim 8, Zhang, Schmitt, Park, Taguchi, and Huber teach: The method as in claim 7 wherein the pre-selected criteria are selected from: user-defined criteria; default criteria; dynamically-determined criteria; a pre-selected density; and combinations thereof (see Huber at least [0024] Down-sampling the point cloud data spatially, such as by minimum distance between points to even out density, sensor noise, and fine features that are not important in large plane detection).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the vehicle localization method disclosed by Zhang, Schmitt, Park, Taguchi, and Huber to include the data down-sampling of Huber. One of ordinary skill in the art would have been motivated to make this modification because down-sampling the data that is used to identify planes from point cloud data allows the best data to be considered and create the most accurate planes, as suggested by Huber (see Huber at least [0023] Increased likelihood of a best possible solution can be achieved by either increasing the number of iterations or, more interestingly, reducing the number of points that are considered by RANSAC to the most likely candidates).
Claim(s) 11, 12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Zhang, in view of Schmitt, further in view of Park, further in view of Taguchi, and further in view of US 20210311504 A1 Chai, Eugene et al. (hereinafter Chai).
Regarding claim 11, Zhang, Schmitt, Park, and Taguchi teach: The method as in claim 9.
Zhang, Schmitt, Park, and Taguchi do not teach: identifying planar points as points belonging to non-ground planes.
Chai teaches: identifying planar points as points belonging to non-ground planes (see Chai at least [0104] reconstruction system 100 can eliminate the ground plane as that plane whose points are consistent with the drone's height).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the vehicle localization method disclosed by Zhang, Schmitt, Park, and Taguchi to include the identification of non-ground planar points of Chai. One of ordinary skill in the art would have been motivated to make this modification because distinguishing between ground planes and non-ground planes allows the unwanted planes to be discarded so computation can be focused on the desired type of planes, as suggested by Chai (see Chai at least [0104] Intuitively, for example, the rooftop would be a large, uniformly oriented surface (surface normal variance is low) that lies above the ground plane... So, it discards all planes that do not satisfy this definition (this includes planes with high variances and the ground surface)).
Regarding claim 12, Zhang, Schmitt, Park, and Taguchi teach: The method as in claim 9 further comprising: identifying planar points (see Taguchi at least [0040] determine inliers on a plane) including:
(a) selecting a random point from downsampled data (see Taguchi at least [0038] we randomly select several reference points 250 q.sub.j,r (r=1, . . . , N) from the inliers of the j th plane landmark);
(d) identifying the non-planar points as not the planar points (see Taguchi at least [0047] the point and plane measurements located as inliers in the RANSAC-based registration are associated to corresponding landmarks, while those located as outliers are discarded).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the image-based vehicle localization method disclosed by Zhang, Schmitt, Park, and Taguchi to include the random point selection and non-planar point identification of Taguchi. One of ordinary skill in the art would have been motivated to make this modification because randomly selecting points to compare between different frames is a less time-consuming method of determining planar features in localization processes, as suggested by Taguchi (see Taguchi at least [0037] Instead of performing a time-consuming plane extraction procedure on each frame independently from other frames, as is the prior art, we make use of the predicted pose to extract planes).
Zhang, Schmitt, Park, and Taguchi do not teach: (b) locating point neighbors of the random point; (c) identifying the planar points in the downsampled data as points that lie on a planar plane formed by the random point and the point neighbors, if any; and (e) repeating steps (a)-(d) until no downsampled data remain to be examined, or a pre-selected number of the planar planes has been achieved, or a pre-selected number of iterations of steps (a)-(d) has been executed.
Chai teaches: (b) locating point neighbors of the random point (see Chai at least [0105] reconstruction system 100 forms clusters of points based on their spatial relationships such that neighboring points belong to the same cluster);
(c) identifying the planar points in the downsampled data as points that lie on a planar plane formed by the random point and the point neighbors, if any (see Chai at least [0103] reconstruction system 100 can use a plane-fitting algorithm (e.g., a plane extraction (PE) method based on the random sample consensus (RANSAC) approach) to segment the LIDAR points into groups of points that fall onto planes. In each plane, reconstruction system 100 removes outlying points that are further away from neighboring points in the same plane using a statistical outlier filter); and
(e) repeating steps (a)-(d) until no downsampled data remain to be examined, or a pre-selected number of the planar planes has been achieved, or a pre-selected number of iterations of steps (a)-(d) has been executed (see Chai at least [0104] At the end of the computation and analysis, reconstruction system 100 classifies a single plane as the roof surface).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the vehicle localization and plane identification method disclosed by Zhang, Schmitt, Park, and Taguchi to include the point-clustering and plane-fitting technique of Chai. One of ordinary skill in the art would have been motivated to make this modification because analyzing clusters helps differentiate planar results of varying sizes depending on desired results, as suggested by Chai (see Chai at least [0105] This way, points belonging to different objects form clusters. Since the roof is normally a relatively larger surface, reconstruction system 100 simply discards smaller clusters).
Claim(s) 13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Zhang, in view of Schmitt, further in view of Park, further in view of Taguchi, further in view of Huber, and further in view of US 20200111251 A1 SHI; Wenzhong et al. (hereinafter Shi).
Regarding claim 13, Zhang, Schmitt, Park, Taguchi, and Huber teach: The method as in claim 7.
Zhang, Schmitt, Park, Taguchi, and Huber do not teach: further comprising forming real-time planes that are grown to outliers from a plurality of scans of the downsampled real-time data.
Shi teaches: further comprising forming real-time planes that are grown to outliers from a plurality of scans of the downsampled real-time data (see Shi at least [0070] Regarding the structural element detection, the vertical planes that are classified as wall structures are analyzed and some of them need to be pruned. Due to noise in the point cloud, some vertical planes may not belong to the wall structures. These planes will be pruned using region growth, histogram, and probability analysis. For each wall plane, region growth is applied to remove outliers from each wall plane).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the multiple-scan vehicle localization method disclosed by Zhang, Schmitt, Park, Taguchi, and Huber to include the outlier-inclusive region growth technique of Shi. One of ordinary skill in the art would have been motivated to make this modification because growing regions including outliers may allow the system to recognize planes even if they may have a gap in data due to an obstruction, and then later remove the outliers after determining the plane type, as suggested by Shi (see Shi at least [0088] Predefined thresholds are used to apply region growth algorithm to distinguish different wall voids. Some detected voids may be caused by furniture blocking).
Claim(s) 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Zhang, in view of Schmitt, further in view of Park, further in view of Taguchi, further in view of Huber, further in view of Shi, and further in view of JP 2015108621 A FENG CHEN et al. (hereinafter Feng).
Regarding claim 14, Zhang, Schmitt, Park, Taguchi, Huber, and Shi teach: The method as in claim 13.
Zhang, Schmitt, Park, Taguchi, Huber, and Shi do not teach: determining discontinuities in the real-time planes; and removing points that are part of the discontinuities.
Feng teaches: determining discontinuities in the real-time planes (see Feng at least [pg. 4, para. 10, beginning with “Two other different”] regions at the boundary between the two planes); and
removing points that are part of the discontinuities (see Feng at least [pg. 4, para. 10, beginning with “Two other different”] For this reason, such a region is excluded).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the vehicle localization and plane-identification method disclosed by Zhang, Schmitt, Park, Taguchi, Huber, and Shi to include the removal of discontinuities from otherwise planar point cloud data technique of Feng. One of ordinary skill in the art would have been motivated to make this modification because discontinuities such as boundary regions between two planes can yield misleading results if not removed, as suggested by Feng (see Feng at least [pg. 4, para. 10, beginning with “Two other different”] the boundary regions are close to each other in 3D, but include points on two different planes, such as the corners of the room, which reduces the plane fitting accuracy after merging).
Claim(s) 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Zhang, in view of Schmitt, further in view of Park, further in view of Taguchi, further in view of Chai, further in view of Huber, and further in view of Shi.
Regarding claim 15, Zhang, Schmitt, Park, Taguchi, and Chai teach: The method as in claim 12.
Zhang, Schmitt, Park, Taguchi, and Chai do not teach: forming real-time planes that are grown to outliers from a plurality of scans of the downsampled real-time data.
Shi teaches: forming real-time planes that are grown to outliers from a plurality of scans of the downsampled real-time data (see Shi at least [0070] Regarding the structural element detection, the vertical planes that are classified as wall structures are analyzed and some of them need to be pruned. Due to noise in the point cloud, some vertical planes may not belong to the wall structures. These planes will be pruned using region growth, histogram, and probability analysis. For each wall plane, region growth is applied to remove outliers from each wall plane).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the multiple-scan vehicle localization method disclosed by Zhang, Schmitt, Park, Taguchi, and Chai to include the outlier-inclusive region growth technique of Shi. One of ordinary skill in the art would have been motivated to make this modification because growing regions including outliers may allow the system to recognize planes even if they may have a gap in data due to an obstruction, and then later remove the outliers after determining the plane type, as suggested by Shi (see Shi at least [0088] Predefined thresholds are used to apply region growth algorithm to distinguish different wall voids. Some detected voids may be caused by furniture blocking).
Zhang, Schmitt, Park, Taguchi, Chai, and Shi do not teach: matching the real-time planes with the planar planes.
Huber teaches: matching the real-time planes with the planar planes (see Huber at least [0047] Another exemplary and non-limiting embodiment encompasses the extension of floor level detection and tracking used as a constraint to assist in scan matching while on a single floor. For example, once enough data is captured on a building story, the floor plane may be detected, then floor planes in new frames of scan data may be calculated and corrections may be applied to bring this data into level with the existing floor either to refine the initial guess of fit before scan matching or as a constraint in the scan matching optimization function).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the vehicle localization method disclosed by Zhang, Schmitt, Park, Taguchi, Chai, and Shi to include the plane-matching technique of Huber. One of ordinary skill in the art would have been motivated to make this modification because identifying correspondences between planes helps contextualize environmental data in real-time through known features such as a level floor, as suggested by Huber (see Huber at least [0046] real time correction of floor data while scanning, referred to as "live scanning" may allow one to use an established floor level while scanning to improve the process of scanning in difficult environments).
Claim(s) 16, 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Zhang, in view of Schmitt, further in view of Park, further in view of Taguchi, further in view of Chai, and further in view of US 20140010407 A1 Sinha; Sudipta N et al. (hereinafter Sinha).
Regarding claim 16, Zhang, Schmitt, Park, Taguchi, and Chai teach: The method as in claim 12.
Zhang, Schmitt, Park, Taguchi, and Chai do not teach: wherein organizing the first map comprises creating a k- dimensional tree from the first map and/or creating, by a parallel processor, a plurality of k-dimensional trees from the first map and the second map.
Sinha teaches: wherein organizing the first map comprises creating a k- dimensional tree from the first map and/or creating, by a parallel processor, a plurality of k-dimensional trees from the first map and the second map (see Sinha at least [0072] an out-of-core approach should be possible for larger scenes, where the 3DR map is partitioned into overlapping submaps, kd-trees are built for each of them and only a few relevant sub-maps need to be in the main memory at any one time).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the vehicle localization method disclosed by Zhang, Schmitt, Park, and Taguchi to include the kd-tree map data organization technique of Sinha. One of ordinary skill in the art would have been motivated to make this modification because using kd-trees makes the process of searching and matching features between multiple datasets more efficient, as suggested by Sinha (see Sinha at least [0024] For determining these correspondences, in one embodiment, features are matched using so-called DAISY descriptors, and a kd-tree index to efficiently search the database of indexed DAISY descriptors).
Regarding claim 17, Zhang, Schmitt, Park, Taguchi, and Chai teach: The method as in claim 12.
Zhang, Schmitt, Park, Taguchi, and Chai do not teach: wherein organizing the first map data set and the second map comprises creating a plurality of k-dimensional trees from the first map data set and the second map.
Sinha teaches: wherein organizing the first map data set and the at least one second map comprises creating a plurality of k-dimensional trees from the first map data set and the at least one second map (see Sinha at least [0072] an out-of-core approach should be possible for larger scenes, where the 3DR map is partitioned into overlapping submaps, kd-trees are built for each of them and only a few relevant sub-maps need to be in the main memory at any one time).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the vehicle localization method disclosed by Zhang, Schmitt, Park, and Taguchi to include the kd-tree submap data organization technique of Sinha. One of ordinary skill in the art would have been motivated to make this modification because using kd-trees makes the process of searching and matching features between multiple datasets more efficient, as suggested by Sinha (see Sinha at least [0024] For determining these correspondences, in one embodiment, features are matched using so-called DAISY descriptors, and a kd-tree index to efficiently search the database of indexed DAISY descriptors).
Claim(s) 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Zhang, in view of Schmitt, further in view of Park, further in view of Taguchi, and further in view of CA 3132803 A1 MCNICHOLS JOHN et al. (hereinafter McNichols).
Regarding claim 18, Zhang, Schmitt, Park, and Taguchi teach: The method as in claim 6.
Zhang, Schmitt, Park, and Taguchi do not teach: wherein the border region comprises a width that is dynamically determined based on a speed of the autonomous vehicle, a number and a type of obstacles surrounding the autonomous vehicle, and/or characteristics of an environment surrounding the autonomous vehicle.
McNichols teaches: wherein the border region comprises a width that is dynamically determined based on a speed of the autonomous vehicle, a number and a type of obstacles surrounding the autonomous vehicle, and/or characteristics of an environment surrounding the autonomous vehicle (see McNichols at least [0134] FIG. 14A illustrates an example computer-implemented process of automatically determining dimensions of field buffer regions during agricultural operations based on localized real-time weather conditions).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the vehicle localization method disclosed by Zhang, Schmitt, Park, and Taguchi to include the dynamically-determined buffer regions between different travel areas of McNichols. One of ordinary skill in the art would have been motivated to make this modification because environmental conditions near the vehicle may require different levels of precaution in order to perform tasks according to weather demands, as suggested by McNichols (see McNichols at least [0137] For example, step 1408 may comprise enlarging or reducing the size of a previously defined buffer region based upon the effect calculated at step 1406; high wind speed could require a much larger buffer region to avoid drift of product into the buffer region based on the then-current GPS position of the apparatus 702, and low wind speed could permit observing a smaller buffer region).
Response to Arguments
Applicant's arguments filed 10/08/2025 have been fully considered.
Applicant's amendments and arguments overcome the objections to the specification.
Applicant's amendments overcome the objections to the claims.
Applicant's amendments overcome the 35 U.S.C. §101 rejection for claims 1-18.
Regarding the arguments provided for the 35 U.S.C. §103 rejections of claims 1-18 (remarks pages 14-17), the applicant's arguments have been considered but are moot because of new grounds of rejection.
The argued limitation regarding temporally and positionally dynamic thresholds is taught by newly cited reference Schmitt.
Since claim 1 remains rejected under 35 U.S.C. §103, dependent claims 2-18 cannot be considered allowable through dependency. Claims 2-18 stand rejected as described above.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US 20180089832 A1 Liu; Haowei et al. describes using changing thresholds in the matching process of determining re-localization candidate frames
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ELLE ROSE KNUDSON whose telephone number is (703)756-1742. The examiner can normally be reached 1000-1700 ET M-F.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Hitesh Patel can be reached at (571) 270-5442. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/ELLE ROSE KNUDSON/Examiner, Art Unit 3667
/Hitesh Patel/Supervisory Patent Examiner, Art Unit 3667
1/23/26