US 12,436,261 B1
Radar double bounce detection
Arvind Srivastav, San Francisco, CA (US); Jifei Qian, San Jose, CA (US); and Yifan Zuo, Foster City, CA (US)
Assigned to Zoox, Inc., Foster City, CA (US)
Filed by Zoox, Inc., Foster City, CA (US)
Filed on Apr. 28, 2023, as Appl. No. 18/140,880.
Int. Cl. G01S 13/58 (2006.01); G01S 7/02 (2006.01); G01S 13/72 (2006.01); G01S 13/931 (2020.01)
CPC G01S 13/584 (2013.01) [G01S 7/023 (2013.01); G01S 13/726 (2013.01); G01S 2013/93271 (2020.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
one or more processors; and
one or more non-transitory computer-readable media storing computer-executable instructions that, when executed, cause the one or more processors to perform operations comprising:
receiving, from a radar sensor associated with an autonomous vehicle, radar data of an environment, the radar data including a set of radar points;
inputting the set of radar points into a machine-learning model;
receiving, from the machine-learning model, a first bounding box associated with a first object and a second bounding box associated with a second object;
determining, based at least in part on the first bounding box, a frustrum extending from the radar sensor and encompassing the first bounding box;
determining that a center of the second bounding box is located within the frustrum;
determining a first cluster of the set of radar points associated with the first object and a second cluster of the set of radar points associated with the second object;
determining a first range associated with the first cluster and a second range associated with the second cluster;
determining a first velocity associated with the first cluster and a second velocity associated with the second cluster;
determining, based at least in part on the first range and the second range, that a range difference is within a first threshold range;
determining, based at least in part on the first velocity and the second velocity, that a velocity difference is within a second threshold range;
determining, based at least in part on the center of the second bounding box being within the frustrum, the range difference being within the first threshold range, and the velocity difference being within the second threshold range, that the second cluster of the set of radar points is a false positive object; and
controlling the autonomous vehicle based at least in part on the second cluster being the false positive object.