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 .
Claim Objections
Claims 1, 9 and 10, and therefore claims 2–8 which depend from claim 1 are objected to because of the following informalities: claims 1, 9 and 10 all recite the acronym “SLAM” but do not recite what SLAM stands for. Accordingly, the claims should be amended to instead recite “Simultaneous Localization And Mapping (SLAM).” Appropriate correction is required.
Claims 1, 9 and 10, and therefore claims 2–8 which depend from claim 1 are further objected to because these claims all recite “autonomous mobility” but should instead recite “an autonomous mobility system,” since what is understood to be recited here is not the functionality of autonomous mobility, but a device/system in which autonomous mobility is instantiated/deployed. Appropriate correction is required.
Claim Rejections - 35 USC § 103
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.
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 and 9–10 are rejected under 35 U.S.C. 103 as being unpatentable over Shao et al., WIPO PCT Patent Application Publication No. WO2023283987 (with reference to provided machine English language translation, herein “Shao”) in view of Singandhupe et al., "MCC-EKF for Autonomous Car Security," 2020 Fourth IEEE International Conference on Robotic Computing (IRC), Taichung, Taiwan, 2020, pp. 306-313, doi: 10.1109/IRC.2020.00056 (herein “Singandhupe”).
Regarding claims 1, 9 and 10, with substantive differences between the claims noted in curly brackets {}, deficiencies of Shao noted in square brackets, and claim 1 as exemplary, Shao teaches {a SLAM device that is installed in autonomous mobility equipped with a LiDAR sensor and a camera, the SLAM device comprising processing circuitry to perform: - claim 1 / A non-transitory computer readable medium storing a SLAM attack countermeasure program for a computer that is installed in autonomous mobility equipped with a LiDAR sensor and a camera, the SLAM attack countermeasure program causing the computer to execute: - claim 9 / Autonomous mobility comprising: a LiDAR sensor to perform measurement and generate measurement point cloud data; a camera to capture an image and generate image data (Shao ¶¶1, 18, 6, 23, 28, 34, sensor security detection method, device and storage medium in an unmanned system (autonomous mobility) including a first sensor of a Lidar for which a point cloud is determined, and a second sensor of a camera from which an RGB image is acquired); a SLAM device to estimate a location and generate a map based on the measurement point cloud data (Shao ¶88, Lidar location calculated using a SLAM algorithm, which a PHOSITA would understand to be executed on a SLAM device, using the point cloud, and where the location of the Lidar is noted in a coordinate system relative to the environment (map)); and an autonomous driving system to perform autonomous driving based on the estimated location and the generated map, wherein the SLAM device includes processing circuitry to perform: - claim 10 (Shao ¶109, positioning task of autonomous driving using coordinates projected onto a map)}
point cloud processing of acquiring measurement point cloud data from the LiDAR sensor, generating first three-dimensional point cloud data based on the measurement point cloud data (Shao ¶¶18, 88–19, first group of sensors including a Lidar which obtains point clouds in X Y and Z axis (three-dimensional) through scanning using 64-line laser radar (measurement point cloud data)), and performing SLAM using the first three-dimensional point cloud data so as to estimate a location and generate a map (Shao ¶¶30–32, sensing data for the lidar returning a three-dimensional point cloud data of a detection target including coordinates of its location, where ¶77 teaches the lidar is tracked via SLAM which stands for simultaneous localization and mapping, thus the mapping also occurring with the location calculation);
image processing of acquiring image data from the camera, recognizing a feature object in an image indicated in the image data (Shao ¶¶34 and 136, acquired RBG image from the camera is applied to a deep convolutional neural network (image processing) to determine the coordinates of a target object in the image), calculating a distance from the camera to the feature object as a relative distance, and converting the image data into second [three-dimensional point cloud] data (Shao ¶¶32, 34, and 137–139, the coordinates of the target object are determined with respect to a coordinate system of [0,0] for the vehicle body where the camera is located, therefore, the coordinates representing a distance also); and
attack detection of superimposing [a three-dimensional point cloud indicated in the second three-dimensional point cloud data onto the generated map] so as to calculate a distance from the estimated location on the map to the feature object as a comparative distance (Shao ¶¶153–154, establishing (calculate) a deviation difference between perception data of the camera versus the lidar, using the respective coordinate values (thus distances) of the target object as determined by the lidar and the camera, where ¶¶130 and 192 teach the projection (superimposing) of the three dimensional point cloud of the lidar to the coordinate system of the vehicle), comparing the comparative distance with the relative distance, and when a difference between the comparative distance and the relative distance is a value outside an allowable range, (Shao ¶¶175–188, deviation between the lidar coordinates and the camera coordinates for the feature object is determined as a distance and compared to see if the distance is an element of a range as given in ¶¶184, 217 and 220 if the distance value is outside that range, then the sensor is determined to be abnormal and the sensor having been subjected to external attacks) [performing interpolation for the SLAM using the second three-dimensional point cloud data].
Shao does not explicitly teach where Singandhupe teaches image data into a three-dimensional point cloud (Singandhupe page 308, fig. 1, left column, stereo camera data is converted to a odometry point cloud using a SLAM algorithm called Frame-to-Map in rtabmap), superimposing a three-dimensional point cloud indicated in the second three-dimensional point cloud data onto the generated map (Singandhupe fig. 1, page 308, right column, the odometry (three-dimensional point cloud indicated in the odometry from the from the two SLAM methods respective to stereo camera data and lidar data is fused using the disclosed MCC-EKF algorithm), and performing interpolation for the SLAM using the second three-dimensional point cloud data (Singandhupe page 308, right column, fig. 1, MCC-EKF algorithm rejects attack data on the lidar odometry and updates the odometry (interpolation for the SLAM) using the non-attacked SLAM point cloud data from the stereo camera (second three-dimensional point cloud data)).
Therefore taking the teachings of Shao and Singandhupe together as a whole, it would have been obvious to a person having ordinary skill in the art (herein “PHOSITA”) before the effective filing date of the claimed invention to have modified the image data and further processing disclosed in Shao’s abnormal/under attack sensor detection system to use SLAM methods on image data, fuse the image data with lidar data to produce secure attack free odometry as disclosed in Singandhupe at least because doing so would help make an autonomous system robust against attacks. See Singandhupe page 306 right column.
Regarding claim 2, Shao teaches wherein in the attack detection, the processing circuitry performs the interpolation not when the difference is a value outside the allowable range only once, but when the difference is a value outside the allowable range a predetermined number of times in succession (Shao ¶¶175–184, the range in which the distance value is considered to be not abnormal is determined in view of an expected variance which considers an m2 number of normal samples, therefore one sample being outside the allowable range would be within the expected variance, and thus a certain number of abnormal values would need to occur (based on the calculated expected variance) before the distance value falls outside of the range).
Regarding claim 3, Shao does not explicitly teach where Singandhupe teaches wherein in the attack detection, the processing circuitry performs, as the interpolation, estimation of the location by visual odometry and updating of the map using the second three-dimensional point cloud data (Singandhupe page 308, fig. 1, odometry calculated by stereo camera data (visual odometry) and per algorithm 1, the updating of the odometry from MCC-EKF SLAM (map) is using the stereo odometry So).
Therefore taking the teachings of Shao and Singandhupe together as a whole, it would have been obvious to a PHOSITA before the effective filing date of the claimed invention to have modified the image data and further processing disclosed in Shao’s abnormal/under attack sensor detection system to use SLAM methods on stereo odometry to update the odometry when lidar is under attack as disclosed in Singandhupe at least because doing so would help make an autonomous system robust against attacks. See Singandhupe page 306 right column.
Regarding claim 4, with deficiencies of Shao noted in square brackets, Shao teaches wherein in the attack detection, the processing circuitry performs the interpolation when the comparative distance [cannot be] calculated and thus the comparative distance [cannot be] compared with the relative distance (Shao ¶¶153–154, a deviation difference between perception data of the camera versus the lidar, using the respective coordinate values (thus distances) of the target object as determined by the lidar and the camera, where ¶¶130 and 192 teach the projection (superimposing) of the three dimensional point cloud of the lidar to the coordinate system of the vehicle).
While Shao teaches calculating or an attempt to calculate the deviation difference (comparative distance) Shao does not teach a scenario where the comparative difference cannot be calculated and thus cannot be compared. Singandhupe teaches on page 308 that at a time k, a large attack vector ak would cause a large lidar measurement, forcing the lidar odometry value L to zero which would not allow for a comparative difference calculation with another SLAM measurement, and thus have the odometry be updated by the stereo camera data odometry values (performs the interpolation).
Therefore taking the teachings of Shao and Singandhupe together as a whole, it would have been obvious to a PHOSITA before the effective filing date of the claimed invention to have modified the image data and further processing disclosed in Shao’s abnormal/under attack sensor detection system to use SLAM methods on stereo odometry to update the odometry when lidar is under attack and has zero values for its odometry as disclosed in Singandhupe at least because doing so would help make an autonomous system robust against attacks. See Singandhupe page 306 right column.
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Shao in view of Singandhupe as set forth above regarding claim 1 from which claim 8 depends, further in view of Gariepy et al., US Patent No. US 10,585,440 B1 (herein “Gariepy”).
Regarding claim 8, Shao as modified above does not explicitly teach where Gariepy teaches wherein the feature object is a wall (Gariepy col. 12, ll. 50–56, lidar is used to measure the distance between the vehicle and an object such as a wall).
Therefore taking the teachings of Shao and Singandhupe together as a whole, it would have been obvious to a PHOSITA before the effective filing date of the claimed invention to have modified the image data and further processing disclosed in Shao’s abnormal/under attack sensor detection system to use a wall as an object reference as disclosed in Gariepy at least because doing so would be using a commonly found reference known to be within an environment that an autonomous vehicle moves and thus assist in collision avoidance. See Gariepy col. 12, ll. 50–53, col. 10, ll. 49–54, and col. 2, ll. 51–55.
Allowable Subject Matter
Claims 5–7 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims. The following is a statement of reasons for the indication of allowable subject matter:
Regarding claim 5, the closest cited art of record includes Shao and Singandhupe as applied above for claim 1. While Shao in view of Singandhupe teach or suggest reliance on the image based odometry when lidar is under attack, neither reference discloses switching operation to a simplified operation where image processing and attack detection are not performed, in combination with all of the other limitations recited in claims 5 and 1.
Regarding claim 6, the closest cited art of record includes Shao and Singandhupe as applied above for claim 1. While Shao in view of Singandhupe at least suggests scenarios where the comparative distance is not calculated, neither Shao nor Singandhupe teaches or suggests stopping the SLAM device when the comparative distance is not calculated, in combination with all of the other limitations recited in claims 6 and 1.
Regarding claim 7, the closest cited art of record includes Shao and Singandhupe as applied above for claim 1. While Shao in view of Singandhupe teaches sensors alternative to the Lidar sensor, neither Shao nor Singandhupe teaches or suggests an operation mode where the alternative sensor is used in place of the lidar sensor, in combination with all of the other limitations recited in claims 7 and 1.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Labbé. (2019). RTAB‐Map as an open‐source lidar and visual simultaneous localization and mapping library for large‐scale and long‐term online operation. Journal of Field Robotics., 36(2). https://doi.org/10.1002/rob.21831. Labbe is directed towards the RTAB-Map open source library which provides visual odometry including SLAM for image data.
Sun et al., "Towards Robust LiDAR-based Perception in Autonomous Driving: General Black-box Adversarial Sensor Attack and Countermeasures," Proceedings of the 29th USENIX Security Symposium. August 12–14, 2020, pp. 877-894. Sun is directed towards sensor attack and countermeasures in lidar based autonomous driving systems.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MICHELLE M KOETH whose telephone number is (571)272-5908. The examiner can normally be reached Monday-Thursday, 09:00-17:00, Friday 09:00-13:00, EDT/EST.
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, Vincent Rudolph can be reached at 571-272-8243. 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.
MICHELLE M. KOETH
Primary Examiner
Art Unit 2671
/MICHELLE M KOETH/Primary Examiner, Art Unit 2671
Prosecution Insights