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 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. There is a total of 19 claims and claims 1-4, 6-16 and 18-21 are pending.
Response to Amendment
Applicant's argument, filed on August 21, 2026, has been entered and carefully considered. Claims 1-4, 6-11, 13-16 and 20-21 are amended and claims 5 and 17 have been canceled. Claims 1-4,6-16 and 18-21 are pending.
Response to Arguments
Applicants’ arguments filed on 08/21/2026 remarks have been fully considered but are in moot in view of the new ground(s) of rejection which is deemed appropriate to address all of the needs at this time.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-4, 6-16 and 18-21 are rejected under 35 U.S.C. 103 as being unpatentable over Xinyu et al. (NPL-Analyzing Infrastructure LiDAR Placement) .in view of Hanjiang et al.(NPL: Investigating the Impact of Multi-LiDAR Placement on Object Detection for Autonomous Driving; Pub. May 04, 2022; given by applicant in the IDS).
Regarding claim 1, Xinyu discloses a method of automatically determining sensor placement in a target environment comprising a road traffic environment, the method comprising([see in Fig. 1, 6 and abstract]- optimal placement of infrastructure sensors is rarely studied. In this paper, we investigate the problem of infrastructure sensor placement and propose a pipeline that can efficiently and effectively find optimal installation positions for infrastructure sensors in a realistic simulated environment. To better simulate and evaluate LiDAR placement, we establish a Realistic LiDAR; [Fig. 6 ]- in fig. 6 discloses three common traffic scenarios in the test set as examples for evaluating infrastructure LiDAR placements, as shown in Figure (. To visualize the point cloud distribution in InfraLOB, we project the point cloud data into the world coordinate system. Scenario A is the T-junction of urban roads. We place one, two, and three LiDARs at the T-junction to analyze how the number of LiDARs will affect the perception accuracy. It can be observed that more LiDAR brings higher Infra D and lower InfraNUC, which also leads to performance gain. Scenario B is the intersection of urban roads.); a number of a plurality of light detection and ranging (LiDAR) sensors to be mounted on roadside infrastructure in the road traffic environment([see in Fig. 1, 6 and abstract]- optimal placement of infrastructure sensors is rarely studied. In this paper, we investigate the problem of infrastructure sensor placement and propose a pipeline that can efficiently and effectively find optimal installation positions for infrastructure sensors in a realistic simulated environment. To better simulate and evaluate LiDAR placement, we establish a Realistic LiDAR; [Fig. 6 ]- in fig. 6 discloses three common traffic scenarios in the test set as examples for evaluating infrastructure LiDAR placements, as shown in Figure (. To visualize the point
cloud distribution in InfraLOB, we project the point cloud data into the world coordinate system. Scenario A is the T-junction of urban roads. We place one, two, and three
LiDARs at the T-junction to analyze how the number of LiDARs will affect the perception accuracy. It can be observed that more LiDAR brings higher Infra D and lower InfraNUC, which also leads to performance gain. Scenario B is the intersection of urban roads.); a set of placement configuration parameters for the plurality of LiDAR sensors comprising position, height and angle placement configuration parameters for each of the LiDAR sensors on the roadside infrastructure ([pg. 1, right column; first paragraph]- the LiDAR perception problem from the viewpoint of LiDAR placement [12], [13], which is a new perspective and crucial since improper LiDAR placements may cause poor-quality sensing data and thus lead to inferior performance of perception. With the rapid development of V2X applications, knowing how to choose the optimal placement to maximize the benefits brought by infrastructure sensors is necessary. However, existing research on LiDAR placement mainly focuses on vehicles and ignores the placement problem for infrastructure sensors. The placement of infrastructure LiDAR has a higher degree of freedom – both the xyz positions and roll, pitch, and yaw angles need to be considered. Therefore, in this paper, we aim to analyze the interplay between infrastructure LiDAR sensor placement and perception performance); and a set of constraints for the plurality of LiDAR sensors in the road traffic environment([pg. 2, col. 2, third paragraph(LiDAR placement)]- LiDAR placement. Previous research on LiDAR placement mainly focused on the optimization placement of Surround LiDAR for autonomous driving. To tackle the surrounding LiDAR placement problem for autonomous driving, Ma et al. [13] proposed a new method based on Bayesian theory conditional entropy to evaluate vehicle sensor configuration).
However, Xinyu does not explicitly discloses receiving as inputs: a three-dimensional (3D) map of the road traffic environment; based on the received inputs, using a simulation platform to simulate, based on one or more defined conditions in the road traffic environment, operation of the plurality of LiDAR sensors to generate a dataset comprising simulated output data for the plurality of LiDAR sensors; and using a reinforcement learning (RL) algorithm to determine from the dataset comprising simulated output data for the plurality of LiDAR sensors an optimized or at least an improved set of placement configuration parameters for the plurality of LiDAR sensors, wherein the method is applied iteratively for a plurality of adjustments to the received inputs to thereby determine the optimized or at least the improved set of placement configuration parameters for the plurality of LiDAR sensors in the road traffic environment.
In an analogous art, Hanjiang discloses receiving as inputs: a three-dimensional (3D) map of the road traffic environment([see in Fig. 5]-receiving a 3D map); based on the received inputs, using a simulation platform to simulate, based on one or more defined conditions in the road traffic environment, operation of the plurality of LiDAR sensors to generate a dataset comprising simulated output data for the plurality of LiDAR sensors([see in 4.1. Experimental Setup]-Given CARLA scenarios and the target object, we obtain the POG based on the bounding box labels and calculate the surrogate metric S-MIG for every LiDAR placement. For each evaluated LiDAR configuration candidate, we first collect point cloud data in CARLA, then train and test all the object detection models using the collected data. Finally, we correlate the surrogate metric and the 3D detection performance); and using a reinforcement learning (RL) algorithm to determine from the dataset comprising simulated output data for the plurality of LiDAR sensors an optimized or at least an improved set of placement configuration parameters for the plurality of LiDAR sensors([abstract, section:3-Methodology and section 4.2]- LiDAR placement influence on 3D object detection. From Table and Figure we present the 3D detection (3D IoU) and bird-eye-view (BEV) performance of representative voxel-based and point-based detection algorithms using point cloud collected with different LiDAR configurations. Note that we use the most rigorous detection metrics in our experiments from It can be seen that different LiDAR placements clearly influence the detection performance for all the algorithms; [see also in Introduction section: ]-LiDAR sensors); wherein the method is applied iteratively for a plurality of adjustments to the received inputs to thereby determine the optimized or at least the improved set of placement configuration parameters for the plurality of LiDAR sensors in the road traffic environment([pg. 13, left column, second paragrah]- there are some detection algorithms where the fluctuation caused by LiDAR placement is even more significant than the difference between different algorithms given any LiDAR placement. Specifically, the recall with 0.7 IoU of Second is better than PointRCNN under Pyramid-roll, while Second performs worse than Point RCNN using data collected under other LiDAR placements, showing that LiDAR configuration is also a critical factor in object detection. Therefore, there is still room to improve the 3D detection performance from LiDAR placement). Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide the technique of Hanjiang to the modified system of Xinyu a new data collection, detection model training and evaluation framework in the realistic CARLA simulator to evaluate disparate multi-LiDAR configurations. Using several prevalent placements inspired by the designs of set-driving companies, show the correlation between our surrogate metric and object detection performance of different representative algorithms on KITTI through extensive experiments, validating the effectiveness of our LiDAR placement evaluation approach. Results show that sensor placement is non-negligible in 3D point cloud-based object detection, which will contribute up to 10% performance discrepancy in terms of average precision in challenging 3D object detection settings. We believe that this is one of the first studies to quantitatively investigate the influence of LiDAR placement on perception performance [Hanjiang; abstract].
Regarding claim 2, Hanjiang discloses using the simulation platform to generate respective datasets for a plurality of road traffic environment scenarios([see in 4.1. Experimental Setup]-Given CARLA scenarios and the target object, we obtain the POG based on the bounding box labels and calculate the surrogate metric S-MIG for every LiDAR placement. For each evaluated LiDAR configuration candidate, we first collect point cloud data in CARLA, then train and test all the object detection models using the collected data. Finally, we correlate the surrogate metric and the 3D detection performance); and using the respective datasets to train the RL algorithm to determine optimized or at least improved sets of placement configuration parameters for the plurality of sensors for the plurality of target environment scenarios([abstract, section:3-Methodology and section 4.2]- LiDAR placement influence on 3D object detection. From Table and Figure, we present the 3D detection (3D IoU) and bird-eye-view (BEV) performance of representative voxel-based and point-based detection algorithms using point cloud collected with different LiDAR configurations. Note that we use the most rigorous detection metrics in our experiments from It can be seen that different LiDAR placements clearly influence the detection performance for all the algorithms).
Regarding claim 3, Hanjiang discloses wherein one of the received inputs includes an optimized or at least an improved set of placement configuration parameters for one of the plurality of road traffic environment scenarios environment ([pg. 13, left column, second paragrah]- there are some detection algorithms where the fluctuation caused by LiDAR placement is even more significant than the difference between different algorithms given any LiDAR placement. Specifically, the recall with 0.7 IoU of Second is better than PointRCNN under Pyramid-roll, while Second performs worse than Point RCNN using data collected under other LiDAR placements, showing that LiDAR configuration is also a critical factor in object detection. Therefore, there is still room to improve the 3D detection performance from LiDAR placement)
Regarding claim 4, Hanjiang discloses using data generated for the respective datasets for the plurality of road traffic environment scenarios to dynamically adapt to new or different road traffic environment scenarios or scenes ([Section 4]- CARLA simulator. Use CARLA as the simulation platform to collect data and evaluate our method. CARLA is a high-definition open-source simulator for autonomous driving research [ that offers flexible scenario setups and sensor configurations. CARLA provides realistic rendering and physics simulation based on Unreal Engine 4. Furthermore, the LiDAR sensor could generate accurate point cloud data and could be easily configured with customized placement).
Claim 5(canceled).
Regarding claim 6, Hanjiang discloses wherein the method is applied iteratively for a plurality of LiDAR sensor model types, the LiDAR sensor model types having different values for the constraints in the road traffic environment, to thereby determine an optimized or at least an improved set of placement configuration parameters for the plurality of LiDAR sensors including LiDAR sensor model type([see in Table-1,2 and table-5]-Table-1 discloses LiDAR placement influence on 3D object detection. From Table and Figure we present the 3D detection (3D IoU) and bird-eye-view (BEV) performance of representative voxel-based and point-based detection algorithms using point cloud collected with different LiDAR configurations. Note that we use the most rigorous detection metrics in our experiments from It can be seen that different LiDAR placements clearly influence the detection performance for all the algorithms, varying 10% at most ).
Regarding claim 7, Hanjiang discloses wherein a reward parameter for the RL algorithm comprises simulated coverage area for each of the plurality of LiDAR sensors([abstract, section:3-Methodology and section 4.2]- LiDAR placement influence on 3D object detection. From Table and Figure we present the 3D detection (3D IoU) and bird-eye-view (BEV) performance of representative voxel-based and point-based detection algorithms using point cloud collected with different LiDAR configurations. Note that we use the most rigorous detection metrics in our experiments from It can be seen that different LiDAR placements clearly influence the detection performance for all the algorithms).
Regarding claim 8, Hanjiang discloses wherein the method comprises maximizing LiDAR sensor scan coverage areas to converge towards optimal LiDAR sensor placement but taking into account a hybrid objective comprising any one or more of: budget for LiDAR sensor placement; LiDAR sensor specification(s); constraints on sensor placement; any defined conditions for the road traffic environment ([section: 5]- validating the correlation between perception performance and surrogate metric of LiDAR configuration through representative 3D object detection algorithms. Research in this paper sets a precedent for future work to optimize the placement of multiple LiDARs and co-design the sensor placement and perception algorithms).
Regarding claim 9, Hanjiang discloses wherein the simulated coverage area for each of the plurality of LiDAR sensors comprises simulated 3D point cloud data for each of the plurality of LiDAR sensors([see in Fig. 7]- Point cloud data collected from different LiDAR configurations under the same scenarios with car (in black 3D bounding box) and cyclist (in red 3D bounding box)).
Regarding claim 10, Hanjiang discloses wherein the simulated 3D point cloud data for each of the plurality of LiDAR sensors is approximated based on point distance, point cloud density, and point distribution uniformity([see in Fig. 7 and Table-3]- From Figure we can find that point cloud collected through Center is more uniformly distributed so that cyclists get more points compared to Lineroll, while points on large objects may get saturated so the evaluation with S-MIG may be affected).
Regarding claim 11, Hanjiang discloses wherein the 3D point cloud data for each of the plurality of LiDAR sensors is approximated by: representing the 3D point cloud data as a 2D grid; and applying the L1 Norm (Manhattan Distance) to each data point in 2D grid([pg. 2, right hand column, para 1]- 2D grid based methods project the point cloud data onto a 2D bird eye-view (BEV) to generate bounding boxes. Another way to deal with point cloud is to use 3D voxels and 3D CNN, which is also called voxel-based methods).
Regarding claim 12, Hanjiang discloses wherein the method comprises: dividing the 2D grid into rectangular boxes; assigning "x" and "y" coordinates to each data point in the rectangular boxes; and applying the L1 Norm (Manhattan Distance) to each data point in the rectangular boxes([pg. 2, right hand column, para 1 and Fig. 2]- 2D grid based methods project the point cloud data onto a 2D bird eye-view (BEV) to generate bounding boxes. Another way to deal with point cloud is to use 3D voxels and 3D CNN, which is also called voxel-based methods).
Regarding claim 13, Hanjiang discloses wherein the method comprises, in one or more iterations, reducing the number of the plurality of LiDAR sensors to be placed in the road traffic environment and determining an optimized or at least an improved set of placement configuration parameters for the reduced number of LiDAR sensors([see in Table-1,2 and table-5]-Table-1 discloses LiDAR placement influence on 3D object detection. From Table and Figure we present the 3D detection (3D IoU) and bird-eye-view (BEV) performance of representative voxel-based and point-based detection algorithms using point cloud collected with different LiDAR configurations. Note that we use the most rigorous detection metrics in our experiments from It can be seen that different LiDAR placements clearly influence the detection performance for all the algorithms, varying 10% at most ).
Regarding claim 14, Hanjiang discloses wherein the method comprises inputting the set of placement configuration parameters for the plurality of LiDAR sensors as an initial set of placement configuration parameters and iteratively implementing the method for respective adjustments of one or more of the placement configuration parameters to determine an optimized or at least an improved set of placement configuration parameters for the plurality of LiDAR sensors([pg. 13, left hand column, first para]- there are some detection algorithms where the fluctuation caused by LiDAR placement is even more significant than the difference between different algorithms given any LiDAR placement. Specifically, the recall with 0.7 IoU of second is better than Point RCNN under Pyramid-roll, while second performs worse than Point RCNN using data collected under other LiDAR placements, showing that LiDAR configuration is also a critical factor in object detection. Therefore, there is still room to improve the 3D detection performance from LiDAR placement).
Regarding claim 15, Hanjiang discloses wherein the initial set of placement configuration parameters are selected to provide a semi-optimal direction for the RL model where said initial set of placement configuration parameters are mathematically calculated using trigonometry from data defining the road traffic environment or a road traffic environment scenario([pg. 13, left hand column, first para]- there are some detection algorithms where the fluctuation caused by LiDAR placement is even more significant than the difference between different algorithms given any LiDAR placement. Specifically, the recall with 0.7 IoU of second is better than Point RCNN under Pyramid-roll, while second performs worse than Point RCNN using data collected under other LiDAR placements, showing that LiDAR configuration is also a critical factor in object detection. Therefore, there is still room to improve the 3D detection performance from LiDAR placement).
Regarding claim 16, Hanjiang discloses wherein one or more weighting values applied to one or more of the constraints in the set of constraints for the plurality of LiDAR sensors in the road traffic environment ([see in table-5]-a different values are applied).
Claim 17(canceled).
Regarding claim 18, Hanjiang discloses wherein the simulation platform comprises any of: a CARLA simulator for autonomous driving research; an Autoware (Gazebo) simulator; an Airsim (UE4 & Unity) simulator; or a TORCS (Open GL) simulator([abstract]-CARLA simulator).
Regarding claim 19, Hanjiang discloses wherein the RL algorithm comprises a universal RL model([abstract]- an easy-to-compute information-theoretic surrogate metric to quantitatively and fast evaluate LiDAR placement.for 3D detection of different types of objects. We also present a new data collection, detection model training and evaluation framework in the realistic CARLA simulator to evaluate disparate multi-LiDAR configurations).
Regarding claim 20, the claim is interpreted and rejected for the same reason as set forth in claim 1. Hence; all limitations for apparatus claim 20 have been met in method claim 1.
Regarding claim 21, the claim is interpreted and rejected for the same reason as set forth in claim 1. Hence; all limitations for claim 21 have been met in method claim 1.
.
Citation of Pertinent Prior Art
The prior art are made of record and not relied upon but considered pertinent to applicant’s disclosure:
1. DIMITROV V, US 2021/0389466 A1, discloses receiving a first point cloud from a first light detection and ranging (LiDAR) sensor installed at a first location behind a grille of a vehicle through a processing device (106) and a memory (103), where the first point cloud represents a scene in a front of the vehicle.
2. Karafin et. al., US 2024/0151896 A1, discloses to generation of holographic content comprising sensory information, and more specifically to generation of holographic content from non-holographic information.
3. Esna Ashari Esfahani, US 2023/0376832 A1, discloses a reinforcement learning agent to calibrate one or more vehicle parameters within a virtual environment.
4. Kourous-Harrigan, US 2023/0138084 A1, discloses a three-dimensional grid model of a traffic scene can be determined based on grid element.
5. Malboubi et al., US 2023/0062010 A1, discloses electronic communications, e.g., to an intelligent support framework usable for enhancing a responder network.
Conclusion
THIS ACTION IS MADE FINAL. 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 MD NAZMUL HAQUE whose telephone number is (571)272-5328. The examiner can normally be reached IFW.
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, David Czekaj can be reached at 5712727327. 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.
/MD N HAQUE/Primary Examiner, Art Unit 2487
