CPC G01S 7/4972 (2013.01) [G01S 17/42 (2013.01); G01S 17/931 (2020.01)] | 20 Claims |
1. A computer-implemented method of identifying miscalibrations in light detection and ranging (LiDAR) sensors in an autonomous driving vehicle (ADV), the method comprising the operations of:
calculating an initial number of overlapping cloud points between first point cloud data from a first LiDAR sensor and second point cloud data from a second LiDAR sensor, wherein each of the first LiDAR sensor and the second LiDAR sensor is mounted on the ADV that is in motion;
replacing a set of existing calibration parameters of the second LiDAR sensor with a plurality of sets of recalibration parameters;
for each set of the plurality of sets of recalibration parameters, calculating a revised number of overlapping cloud points between the first point cloud data from the first LiDAR sensor and the second point cloud data from the second LiDAR sensor;
comparing the initial number of overlapping cloud points with each revised number of overlapping cloud points;
identifying a potential miscalibration in at least one of the first LiDAR sensor or the second LiDAR sensor when the initial number of overlapping cloud points is smaller than at least one revised number of overlapping cloud points; and
displaying a notification message on a display screen to indicate that at least one of the first LiDAR sensor or the second LiDAR sensor needs to be recalibrated.
|