Sensor calibration via extrinsic scanning

Patent metadata

Field	Value
Publication number	US-2022050934-A1
Application number	US-202016992231-A
Country	US
Kind code	A1
Filing date	Aug 13, 2020
Priority date	Aug 13, 2020
Publication date	Feb 17, 2022
Grant date	—

Abstract

The subject technology provides solutions for performing extrinsic sensor calibration for vehicle sensors, such as environmental sensors deployed in an autonomous vehicle (AV) context. In some aspects, the disclosed technology relates to a sensor localization system that is configured to: perform a scan, using a 3D scanner, to collect surface data associated with an autonomous vehicle (AV), analyze the surface data to identify a coordinate origin of the AV, and calculate a position of at least one or more AV sensors based on the surface data. Methods and computer-readable media are also provided.

First claim

1 . A sensor localization system, comprising:
one or more processors; a three-dimensional (3D) scanner coupled to the one or more processors; and a computer-readable medium coupled to the processors, the computer-readable medium comprising instructions stored therein, which when executed by the processors, cause the processors to perform operations comprising:
performing a scan, using the 3D scanner, to collect surface data associated with an autonomous vehicle (AV);
analyzing the surface data to identify a coordinate origin of the AV; and
calculating a position of at least one or more AV sensors based on the surface data by determining a coordinate location of the at least one of the one or more AV sensors based on the coordinate origin of the AV.

2 . (canceled)

3 . The sensor localization system of claim 1 , wherein analyzing the surface data to identify the coordinate origin for the AV further comprises:
automatically identifying one or more or more known vehicle geometries.

4 . The sensor localization system of claim 1 , wherein performing the scan, further comprises:
receiving a plurality of images of the AV surface; and stitching the plurality of images to generate a three-dimensional image of the AV surface.

5 . The sensor localization system of claim 1 , wherein the surface data further includes computer aided design (CAD) modeling information for the one or more AV sensors.

6 . The sensor localization system of claim 1 , wherein the 3D scanner comprises a stereo camera.

7 . The sensor localization system of claim 1 , wherein the 3D scanner comprises one or more lasers.

8 . The sensor localization system of claim 1 , wherein the one or more AV sensors comprises a Light Detection and Ranging (LiDAR) sensor.

9 . A computer-implemented method comprising:
performing a scan, using a 3D scanner, to collect surface data associated with an autonomous vehicle (AV); analyzing the surface data to identify a coordinate origin of the AV; and calculating a position of at least one of the one or more AV sensors based on the surface data by determining a coordinate location of the at least one of the one or more AV sensors based on the coordinate origin of the AV.

10 . (canceled)

11 . The computer-implemented method of claim 9 , wherein analyzing the surface data to identify the coordinate origin for the AV further comprises:
automatically identifying one or more or more known vehicle geometries.

12 . The computer-implemented method of claim 9 , wherein performing the scan, further comprises:
receiving a plurality of images of the AV surface; and stitching the plurality of images to generate a three-dimensional image of the AV surface.

13 . The computer-implemented method of claim 9 , wherein the surface data further includes computer aided design (CAD) modeling information for the one or more AV sensors.

14 . The computer-implemented method of claim 9 , wherein the 3D scanner comprises a stereo camera.

15 . The computer-implemented method of claim 9 , wherein the 3D scanner comprises one or more lasers.

16 . A non-transitory computer-readable storage medium comprising instructions stored therein, which when executed by one or more processors, cause the processors to perform operations comprising:
performing a scan, using a 3D scanner, to collect surface data associated with an autonomous vehicle (AV); analyzing the surface data to identify a coordinate origin of the AV; and calculating a position of one or more AV sensors based on the surface data by determining a coordinate location of the at least one of the one or more AV sensors based on the coordinate origin of the AV.

17 . (canceled)

18 . The non-transitory computer-readable storage medium of claim 16 , wherein analyzing the surface data to identify the coordinate origin for the AV further comprises:
automatically identifying one or more or more known vehicle geometries.

19 . The non-transitory computer-readable storage medium of claim 16 , wherein performing the scan, further comprises:
receiving a plurality of images of the AV surface; and stitching the plurality of images to generate a three-dimensional image of the AV surface.

20 . The non-transitory computer-readable storage medium of claim 16 , wherein the surface data further includes computer aided design (CAD) modeling information for the one or more AV sensors.

Assignees

• Gm Cruise Holdings Llc

Inventors

• Edwards Zachary
• Tran David
• Jones Matthew

Classifications

• G06T7/80

  Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration · CPC title

• G06T2207/30252

  Vehicle exterior; Vicinity of vehicle · CPC title

• G06T2207/10028

  Range image; Depth image; 3D point clouds · CPC title

• G01S17/931

  of land vehicles · CPC title

• G01S7/4972 Primary

  Alignment of sensor · CPC title