Vehicle positioning method, apparatus and autonomous driving vehicle

Patent metadata

Field	Value
Publication number	US-11953609-B2
Application number	US-202217711339-A
Country	US
Kind code	B2
Filing date	Apr 1, 2022
Priority date	Sep 29, 2021
Publication date	Apr 9, 2024
Grant date	Apr 9, 2024

Abstract

The present disclosure provides a vehicle positioning method, an apparatus and an autonomous driving vehicle, relating to autonomous driving in the technical field of artificial intelligence, which can be applied to high-definition positioning of the autonomous driving vehicle, the method including: if there is no high-definition map in a vehicle, acquiring intermediate pose information of the vehicle based on a global navigation satellite system and/or an inertial measurement unit in the vehicle, and determining the intermediate pose information as global positioning information; acquiring local positioning information; performing fusion processing to the global pose information and the local pose information to obtain fused pose information; performing compensation processing to the fused pose information according to the global attitude angle information and the local attitude angle information to obtain a position of the vehicle.

First claim

What is claimed is:

1. A vehicle positioning method, comprising:
in response to determining that there is no high-definition map in a vehicle, acquiring intermediate pose information of the vehicle based on at least one of a global navigation satellite system and an inertial measurement unit in the vehicle, and determining the intermediate pose information as global positioning information, wherein the global positioning information comprises global pose information and global attitude angle information;
acquiring local positioning information of the vehicle, wherein the local positioning information is positioning information acquired by a positioning apparatus of the vehicle based on a coordinate system with an initial position of the vehicle as an origin point, and the local positioning information comprises local pose information and local attitude angle information, and performing fusion processing on the global pose information and the local pose information to obtain fused pose information; and
performing compensation processing on the fused pose information according to the global attitude angle information and the local attitude angle information to obtain a position of the vehicle;
wherein the global attitude angle information comprises a global roll angle and a global pitch angle, and the local attitude angle information comprises a local roll angle and a local pitch angle;
wherein the performing compensation processing on the fused pose information according to the global attitude angle information and the local attitude angle information to obtain a position of the vehicle comprises:
determining a first compensation parameter according to the global roll angle and the local roll angle, and determining a second compensation parameter according to the global pitch angle and the local pitch angle; and
performing compensation processing on the fused pose information according to the first compensation parameter and the second compensation perameter to obtain the position of the vehicle.

2. The method according to claim 1 , wherein the global pose information comprises previous-frame global pose information and current-frame global pose information, and a previous frame and a current frame are two adjacent frames in images in a traveling process of the vehicle; the performing fusion processing on the global pose information and the local pose information to obtain fused pose information comprises:
determining a first inter-frame relative pose of the vehicle between two adjacent frames according to the previous-frame global pose information and the current-frame global pose information; and
determining a global confidence degree of the global pose information according to the first inter-frame relative pose and the local pose information, and if the global confidence degree reaches a preset global confidence degree threshold, performing fusion processing on the global pose information and the local pose information to obtain the fused pose information, wherein the preset global confidence degree threshold is set by the positioning apparatus based on means of requirements, historical records and tests, and is used to represent a reliability degree of the global pose information.

3. The method according to claim 2 , wherein the determining a global confidence degree of the global pose information according to the first inter-frame relative pose and the local pose information comprises:
determining a second inter-frame relative pose according to the first inter-frame relative pose and the local pose information, wherein the second inter-frame relative pose is used to represent a discrepancy between the global pose information and the local pose information; and
determining the global confidence degree according to the second inter-frame relative pose.

4. The method according to claim 3 , wherein the local pose information comprises previous-frame local pose information and current-frame local pose information; the determining a second inter-frame relative pose according to the first inter-frame relative pose and the local pose information comprises:
determining a third inter-frame relative pose of the vehicle between two adjacent frames according to the previous-frame local pose information and the current-frame local pose information; and
determining a discrepant pose between the first inter-frame relative pose and the third inter-frame relative pose, and determining the discrepant pose as the second inter-frame relative pose.

5. The method according to claim 2 , wherein the performing, if the global confidence degree reaches a preset global confidence degree threshold, fusion processing on the global pose information and the local pose information to obtain the fused pose information comprises:
if the global confidence degree reaches the preset global confidence degree threshold, acquiring a local confidence degree corresponding to the local pose information, and if the local confidence degree reaches a preset local confidence degree threshold, performing fusion processing on the global pose information and the local pose information to obtain the fused pose information.

6. The method according to claim 2 , further comprising:
if the global confidence degree is smaller than the global confidence degree threshold, determining the position of the vehicle according to the local pose information.

7. The method according to claim 1 , wherein the global pose information comprises previous-frame global pose information and current-frame global pose information, and a previous frame and a current frame are two adjacent frames in images in a traveling process of the vehicle; the performing fusion processing on the global pose information and the local pose information to obtain fused pose information comprises:
determining a first inter-frame relative pose of the vehicle between two adjacent frames according to the previous-frame global pose information and the current-frame global pose information; and
determining a global confidence degree of the global pose information according to the first inter-frame relative pose and the local pose information, and if the global confidence degree reaches a preset global confidence degree threshold, performing fusion processing on the global pose information and the local pose information to obtain the fused pose information, wherein the preset global confidence degree threshold is set by the positioning apparatus based on means of requirements, historical records and tests, and is used to represent a reliability degree of the global pose information.

8. The method according to claim 7 , wherein the determining a global confidence degree of the global pose information according to the first inter-frame relative pose and the local pose information comprises:
determining a second inter-frame relative pose according to the first inter-frame relative pose and the local pose information, wherein the second inter-frame relative pose is used to represent a discrepancy between the global pose information and the local pose information; and
determining the global confidence degree according to the second inter-frame relative pose.

9. The method according to claim 8 , wherein the local pose information comprises previous-frame local pose information and current-frame local pose information; the determining a second inter-frame relative pose according to the first inter-frame relative pose and the local pose information comprises:
determining a third inter-frame relative pose of the vehicle between two adjacent frames according to the previous-frame local pose information and the current-frame local pose information; and
determining a discrepant pose between the first inter-frame relative pose an

Assignees

• Beijing Baidu Netcom Sci & Tech Co Ltd

Inventors

• Hou Shenhua
• He Yuzhe
• Peng Liang
• Wan Guowei

Classifications

• G01S19/48 Primary

  by combining or switching between position solutions derived from the satellite radio beacon positioning system and position solutions derived from a further system · CPC title

• G01S19/485

  whereby the further system is an optical system or imaging system · CPC title

• G01S19/49

  whereby the further system is an inertial position system, e.g. loosely-coupled · CPC title

• G01S19/53

  Determining attitude · CPC title

• G01S5/0258

  by combining or switching between measurements derived from different systems · CPC title