Systems and methods for calibration and validation of non-overlapping range sensors of an autonomous vehicle

Patent metadata

Field	Value
Publication number	US-2024142588-A1
Application number	US-202218050026-A
Country	US
Kind code	A1
Filing date	Oct 26, 2022
Priority date	Oct 26, 2022
Publication date	May 2, 2024
Grant date	—

Abstract

Methods and systems for validating LiDAR sensor calibration are disclosed. The methods include collecting sensor data corresponding to a calibration environment from a LiDAR sensor and segmenting a reconstructed point cloud of the calibration environment to generate a plurality of planes. The methods also include determining a normal distance between each of a plurality of point cloud sweeps and one or more of the plurality of planes, determining a validation score based on the normal distances, and validating a calibration of the LiDAR sensor in response to determining that the validation score is less than a threshold. Each point cloud sweep is associated with a collection of LiDAR scans at a collection time stamp.

First claim

What is claimed is:

1 . A method for validating calibration of a LiDAR sensor mounted on a vehicle, the method comprising:
collecting, from the LiDAR sensor, sensor data corresponding to a calibration environment; segmenting a reconstructed point cloud of the calibration environment to generate a plurality of planes; determining a normal distance between each of a plurality of point cloud sweeps and one or more of the plurality of planes, each of the plurality of point cloud sweeps being associated with a collection of LiDAR scans at each of a plurality of collection time stamps; determining, using a plurality of the normal distances, a validation score; and validating a calibration of the LiDAR sensor in response to determining that the validation score is less than a threshold.

2 . The method of claim 1 , further comprising determining that the LiDAR sensor is not calibrated in response to determining that the validation score is not less than the threshold.

3 . The method of claim 1 , further comprising, in response determining that the LiDAR sensor is not calibrated:
performing an action assessment of the vehicle; and causing the vehicle to perform an action.

4 . The method of claim 3 , wherein the action includes recalibrating the LiDAR sensor.

5 . The method of claim 1 , further comprising generating, from the sensor data, the reconstructed point cloud corresponding to the calibration environment.

6 . The method of claim 1 , wherein collecting the sensor data comprises:
rotating a rotating platform in the calibration environment to a plurality of angular positions, the vehicle being mounted on the rotating platform; and collecting, using that LiDAR sensor, at each of the plurality of angular positions, a point cloud sweep.

7 . The method of claim 6 , further comprising determining the plurality of angular positions based on a field of view (FOV) of the LiDAR sensor such that two or more point cloud sweeps collected by the LiDAR sensor share an overlapping view of the calibration environment.

8 . The method of claim 1 , wherein the plurality of planes correspond to one or more calibration targets in the calibration environment.

9 . The method of claim 1 , further comprising, upon validating the calibration of the LiDAR sensor, performing extrinsic calibration validation of the LiDAR sensor with respect to a second LiDAR sensor mounted on the vehicle, wherein a field of view (FOV) of the LiDAR sensor does not overlap with a FOV of the second LiDAR sensor.

10 . The method of claim 9 , wherein performing the extrinsic calibration validation of the LiDAR sensor with respect to the second LiDAR sensor mounted on the vehicle comprises:
collecting, from the second LiDAR sensor, second sensor data corresponding to the calibration environment; generating, by segmenting a second reconstructed point cloud of the calibration environment, a second plurality of planes, the second reconstructed point cloud being generated based on the second sensor data; determining a second normal distance between each of a plurality of points in the sensor data of the LiDAR sensor, and one or more of the second plurality of planes; determining, using a plurality of the second normal distances, a second validation score; and validating the extrinsic calibration of the LiDAR sensor with respect to the second LiDAR sensor in response to determining that the second validation score is less than a second threshold.

11 . A system for validating calibration of a LiDAR sensor, the system comprising:
a LiDAR sensor mounted on a vehicle; at least one processor; and programming instructions stored in a memory and configured to cause the processor to:
collect, from the LiDAR sensor, sensor data corresponding to a calibration environment;
segment a reconstructed point cloud of the calibration environment to generate a plurality of planes;
determine a normal distance between each of a plurality of point cloud sweeps and one or more of the plurality of planes, each of the plurality of point cloud sweeps being associated with a collection of LiDAR scans at each of a plurality of collection time stamps;
determine, using a plurality of the normal distances, a validation score; and
validate a calibration of the LiDAR sensor in response to determining that the validation score is less than a threshold.

12 . The system of claim 11 , further comprising additional programming instructions that are configured to cause the processor to determine that the LiDAR sensor is not calibrated in response to determining that the validation score is not less than the threshold.

13 . The system of claim 11 , further comprising additional programming instructions that are configured to cause the processor to, in response determining that the LiDAR sensor is not calibrated:
perform an action assessment of the vehicle; and cause the vehicle to perform an action.

14 . The system of claim 13 , wherein the action includes recalibrating the LiDAR sensor.

15 . The system of claim 11 , further comprising additional programming instructions that are configured to cause the processor to generate, from the sensor data, the reconstructed point cloud corresponding to the calibration environment.

16 . The system of claim 11 , wherein the instructions to collect the sensor data comprise instructions to:
rotate a rotating platform in the calibration environment to a plurality of angular positions, the vehicle being mounted on the rotating platform; and collect, using that LiDAR sensor, at each of the plurality of angular positions, a point cloud sweep.

17 . The system of claim 16 , further comprising additional programming instructions that are configured to cause the processor to determine the plurality of angular positions based on a field of view (FOV) of the LiDAR sensor such that two or more point cloud sweeps collected by the LiDAR sensor share an overlapping view of the calibration environment.

18 . The system of claim 11 , wherein the plurality of planes correspond to one or more calibration targets in the calibration environment.

19 . The system of claim 11 , further comprising additional programming instructions that are configured to cause the processor to:
perform extrinsic calibration validation of the LiDAR sensor with respect to a second LiDAR sensor mounted on the vehicle, wherein a field of view (FOV) of the LiDAR sensor does not overlap with a FOV of the second LiDAR sensor; and wherein the instructions to perform the extrinsic calibration validation of the LiDAR sensor with respect to the second LiDAR sensor mounted on the vehicle comprise instructions to:
collect, from the second LiDAR sensor, second sensor data corresponding to the calibration environment,
generate, by segmenting a second reconstructed point cloud of the calibration environment, a second plurality of planes, the second reconstructed point cloud being generated based on the second sensor data,
determine a second normal distance between each of a plurality of points in the sensor data of the LiDAR sensor and one or more of the second plurality of planes,
determine, using a plurality of the second normal distances, a second validation score, and
validate the extrinsic calibration of the LiDAR sensor with respect to the second LiDAR sensor in response to determining that the second validation score is less than a second threshold.

20 . A computer program product comprising a non-transitory computer-readable medium that store

Assignees

• Ford Global Tech Llc

Inventors

• Alismail Hatem
• Schoenberg Michael
• Jian Zhenxiang

Classifications

• G01S7/497 Primary

  Means for monitoring or calibrating · CPC title

• G01S17/87

  Combinations of systems using electromagnetic waves other than radio waves · CPC title

• G01S17/931

  of land vehicles · CPC title

• G01S17/89

  for mapping or imaging · CPC title

• G01S7/4972 Primary

  Alignment of sensor · CPC title