US 12,276,752 B2
Identification of spurious radar detections in autonomous vehicle applications
Xu Chen, Livermore, CA (US); Nichola Abdo, Mountain View, CA (US); Ruichi Yu, Mountain View, CA (US); and Chang Gao, Sunnyvale, CA (US)
Assigned to Waymo LLC, Mountain View, CA (US)
Filed by Waymo LLC, Mountain View, CA (US)
Filed on Aug. 16, 2021, as Appl. No. 17/445,129.
Prior Publication US 2023/0046274 A1, Feb. 16, 2023
Int. Cl. G01S 7/41 (2006.01); B60W 60/00 (2020.01); G01S 13/86 (2006.01); G01S 13/931 (2020.01); G06F 18/24 (2023.01); G06F 18/25 (2023.01); G06T 7/20 (2017.01); G06V 20/56 (2022.01)
CPC G01S 7/417 (2013.01) [B60W 60/001 (2020.02); G01S 7/412 (2013.01); G01S 13/867 (2013.01); G01S 13/931 (2013.01); G06F 18/24 (2023.01); G06F 18/253 (2023.01); G06T 7/20 (2013.01); G06V 20/56 (2022.01); B60W 2420/403 (2013.01); B60W 2420/408 (2024.01); B60W 2554/404 (2020.02); G06T 2207/10044 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30252 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
a sensing system of an autonomous vehicle (AV), the sensing system configured to:
obtain a radar data characterizing intensity of radar reflections from an environment of the AV; and
obtain a camera image depicting a region of the environment of the AV; and
a perception system of the AV, the perception system configured to:
identify, based on the radar data, a first candidate object and a second candidate object within the region of the environment of the AV;
process the radar data and the camera image using one or more machine-learning models (MLMs) to determine that the first candidate object is associated with a first probability to be a real object in the environment of the AV and that the second candidate object is associated with a second probability to be a real object;
determine, responsive to the first probability being above a threshold probability, that the first candidate object is a real object in the environment of the AV;
determine, responsive to the second probability being below the threshold probability, that the second candidate object is a potential spurious object in the environment of the AV; and
confirm, using additional radar data and one or more additional camera images, collected at one or more later times, that the second candidate object is a spurious object in the environment of the AV.