US 12,012,126 B2
Calibration based on semantic objects
Nathaniel Jon Kaiser, Foster City, CA (US); Till Kroeger, San Francisco, CA (US); and Elena Stephanie Stumm, San Francisco, CA (US)
Assigned to Zoox, Inc., Foster City, CA (US)
Filed by Zoox, Inc., Foster City, CA (US)
Filed on Dec. 11, 2020, as Appl. No. 17/119,562.
Prior Publication US 2022/0185331 A1, Jun. 16, 2022
Int. Cl. B60W 60/00 (2020.01); B60W 50/00 (2006.01); B60W 50/02 (2012.01); B60W 50/04 (2006.01); G06F 18/24 (2023.01); G06V 20/40 (2022.01); G06V 20/56 (2022.01)
CPC B60W 60/0018 (2020.02) [B60W 50/0205 (2013.01); B60W 50/045 (2013.01); G06F 18/24 (2023.01); G06V 20/41 (2022.01); G06V 20/56 (2022.01); B60W 2050/0083 (2013.01); B60W 2556/45 (2020.02)] 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 first sensor data from a first sensor associated with an autonomous vehicle in an environment;
receiving second sensor data from a second sensor associated with the autonomous vehicle;
receiving semantic map data associated with the environment based at least in part on a number of semantic objects within a threshold distance of the autonomous vehicle;
determining that first sensor data from the first sensor and second sensor data from the second sensor correspond to a semantic object determined based at least in part on the semantic map data;
determining a center point and covariance data associated with semantic object;
determining, based at least in part on calibration data, the first sensor data, and epipolar geometry, a distance between the center point associated with the second sensor data and an epipolar line;
determining, based at least in part on the distance, a calibration parameter associated with at least one of the first sensor or the second sensor; and
controlling the autonomous vehicle based at least in part on the calibration parameter,
wherein the threshold distance is based on at least one of:
a classification type of the semantic objects in the environment,
a frequency of the semantic objects in the environment, or
a location in the semantic map data.