Non-line-of-sight correction for target detection and identification in point clouds
Examples disclosed herein relate to an autonomous driving system in a vehicle having a radar system with a Non-Line-of-Sight (“NLOS”) correction module to correct for NLOS reflections prior to the radar system identifying targets in a path and a surrounding environment of the vehicle, and a sensor fusion module to receive information from the radar system on the identified targets and compare the information received from the radar system to information received from at least one sensor in the vehicle.
1. An autonomous driving system in a vehicle, comprising:
a radar system comprising a Non-Line-of-Sight (“NLOS”) correction module configured to correct for NLOS reflections prior to the radar system identifying targets in a path and a surrounding environment of the vehicle, wherein the NLOS correction module comprises a planar surface identification module configured to locate at least one planar reflecting surface in a field of view of the radar system; and
a sensor fusion module to receive information from the radar system on the identified targets and compare the information received from the radar system to information received from at least one sensor in the vehicle.
2. The autonomous driving system of claim 1 , wherein the radar system further comprises a beam steering meta-structure antenna having an array of radiating meta-structure cells.
3. The autonomous driving system of claim 1 , wherein the at least one sensor comprises at least one of a camera or a lidar.
4. The autonomous driving system of claim 1 , wherein the NLOS correction module is further configured to receive a point cloud from the radar system and generate a corrected point cloud.
5. The autonomous driving system of claim 1 , wherein the NLOS correction module receives a supplemental point cloud from the at least one sensor in the vehicle.
6. The autonomous driving system of claim 1 , wherein the planar surface identification module identifies the at least one planar reflecting surface using the supplemental point cloud.
7. The autonomous driving system of claim 1 , wherein the NLOS correction module further comprises a reflection remapping module configured to remap the NLOS reflections about the at least one planar reflecting surface.
8. The autonomous driving system of claim 4 , wherein the planar surface identification module comprises a Kernel-Based Hough Transform and is further configured to detect the at least one planar reflecting surface from the point cloud.
9. A radar system, comprising:
a radar module having a meta-structure beam steering antenna configured to radiate a plurality of RF beams under control of an antenna controller;
an NLOS correction module configured to correct for NLOS reflections of targets reflecting the plurality of radiated RF beams;
a perception module configured to determine a control action for the antenna controller and identify the targets; and
a transceiver configured to provide radar data from a radar scan to the perception module.
10. The radar system of claim 9 , further comprising an RFIC configured to provide steering angle control of the meta-structure beam steering antenna.
11. The radar system of claim 9 , wherein the NLOS correction module is further configured to receive a point cloud from the radar data and generate a corrected point cloud.
12. The radar system of claim 11 , wherein the perception module identifies the targets from the corrected point cloud.
13. The radar system of claim 9 , wherein the NLOS correction module comprises a planar surface identification module configured to locate at least one planar reflecting surface in a field of view of the radar system.
14. The radar system of claim 13 , wherein the NLOS correction module further comprises a reflection remapping module configured to remap the NLOS reflections about the at least one planar reflecting surface.
15. A method for operating a radar system in an autonomous driving system, the method comprising:
directing a meta-structure antenna to generate RF beams at a plurality of directions in a field of view of the autonomous driving system;
receiving reflected RF beams from targets in the field of view;
generating a point cloud;
correcting for Non-Line-of-Sight (“NLOS”) reflections in the point cloud;
generating a corrected point cloud, wherein the correcting comprises locating at least one planar reflecting surface in the field of view; and
identifying at least one target from the corrected point cloud.
16. The method of claim 15 , wherein directing the meta-structure antenna to generate the RF beams at the plurality of directions comprises generating the RF beams at a plurality of phase shifts from an RFIC coupled to the meta-structure antenna.
17. The method of claim 15 , wherein correcting for NLOS reflections in the point cloud to generate the corrected point cloud comprises remapping the NLOS reflections about the at least one planar reflecting surface.
18. The radar system of claim 9 , wherein the NLOS correction module receives a supplemental point cloud from at least one sensor in the vehicle.
19. The method of claim 15 , further comprising:
receiving a supplemental point cloud from the at least one sensor in the autonomous driving system.
20. The method of claim 19 , wherein locating the at least one planar reflecting surface in the field of view comprises identifying the at least one planar reflecting surface using the supplemental point cloud.