Method for fusing road data to generate a map, electronic device, and storage medium

Patent metadata

Field	Value
Publication number	US-12498248-B2
Application number	US-202218063772-A
Country	US
Kind code	B2
Filing date	Dec 9, 2022
Priority date	Dec 10, 2021
Publication date	Dec 16, 2025
Grant date	Dec 16, 2025

Abstract

A method for fusing road data to generate a map, includes: determining benchmark road data and at least one road data to be fused in a target road area; establishing, successively for each road data to be fused, a first road element association relationship between the first sub road data and the benchmark road data; establishing a second road element association relationship between the second sub road data and the benchmark road data according to the first road element association relationship; and fusing the benchmark road data and the road data to be fused according to the above association relationships to update the benchmark road data.

First claim

What is claimed is:

1 . A method for fusing road data to generate a map, comprising:
determining benchmark road data and at least one road data to be fused in a target road area, wherein the benchmark road data and the at least one road data to be fused are data collected by a sensor or data collected using a crowdsourcing map, wherein each of the at least one road data to be fused comprises trajectory and surrounding road element data of trajectory points in the trajectory, and the trajectory is obtained by locating with a positioning sensor; obtaining first sub road data and second sub road data by segmenting the trajectory of the road data to be fused according to a positioning signal quality of the positioning sensor for locating the trajectory, wherein a positioning signal quality of first sub road data is higher than that of second sub road data, and the first sub road data comprises a first trajectory segment and surrounding road element data of trajectory points in the first trajectory segment, the second sub road data comprises a second trajectory segment and surrounding road element data of trajectory points in the second trajectory segment, and the first trajectory segment is adjacent to the second trajectory segment, wherein the surrounding road element data comprises world coordinate information of surrounding road elements, and the surrounding road elements comprises at least one of a lane line or a road traffic sign; establishing, successively for each road data to be fused, a first road element association relationship between the first sub road data and the benchmark road data according to the surrounding road element data; establishing a second road element association relationship between the second sub road data and the benchmark road data according to the first road element association relationship; fusing the benchmark road data and the road data to be fused according to the first road element association relationship and the second road element association relationship to update the benchmark road data, until all of the at least one road data to be fused is processed, to obtain road area data of the target road area; generating an autonomous driving map of the target road area according to the road area data, and sending the autonomous driving map to an autonomous driving vehicle for autonomous driving, wherein the autonomous driving map comprises the lane line, and the autonomous driving comprises autonomous driving based on the lane line of the autonomous driving map; and providing the autonomous driving map to the autonomous driving vehicle and causing the vehicle to perform autonomous driving based on the lane line represented in the autonomous driving map, wherein establishing the second road element association relationship between the second sub road data and the benchmark road data according to the first road element association relationship comprises: establishing, by starting from a first end of the second trajectory segment that is adjacent to the first trajectory segment and successively for each of the trajectory points to be processed in the second trajectory segment, a sub association relationship between the surrounding road element data of the trajectory point and the road element data in the benchmark road data according to the first road element association relationship of the first sub road data and the surrounding road elements; and generating the second road element association relationship according to the respective sub association relationships.

2 . The method according to claim 1 , wherein the surrounding road element data comprises feature data of each of surrounding road elements, and establishing, successively for each road data to be fused, the first road element association relationship between the first sub road data and the benchmark road data comprises:
selecting, successively for each road data to be fused and for each surrounding road element in the respective road data to be fused, a road element associated with the surrounding road element from the benchmark road data according to the feature data of the surrounding road element; and generating the first road element association relationship according to the respective surrounding road elements in the road data to be fused and the associated road elements.

3 . The method according to claim 1 , wherein establishing the second road element association relationship between the second sub road data and the benchmark road data according to the first road element association relationship comprises:
determining adjacent first sub road data of the second sub road data, wherein a first trajectory segment in the adjacent first sub road data is adjacent to a second trajectory segment in the second sub road data; establishing, by starting from a first end of the second trajectory segment that is elose adjacent to the first trajectory segment and successively for each of the trajectory points to be processed in the second trajectory segment, a sub association relationship between the surrounding road element data of the trajectory point and the road element data in the benchmark road data according to the first road element association relationship of the adjacent first sub road data; determining, when the respective established sub association relationships meet a preset deviation condition and not all of the trajectory points in the second trajectory segment are processed, an intermediate transformation relationship between the second sub road data and the benchmark road data according to the respective established sub association relationships, and transforming the second sub road data based on the intermediate transformation relationship to obtain updated second sub road data; repeating the above two steps until all of the trajectory points in the second trajectory segment are processed; and generating the second road element association relationship according to the respective established sub association relationships.

4 . The method according to claim 3 , wherein the number of the adjacent first sub road data is one or two.

5 . The method according to claim 1 , wherein fusing the benchmark road data and the road data to be fused according to the first road element association relationship and the second road element association relationship to update the benchmark road data comprises:
determining a first transformation relationship from the first sub road data to the benchmark road data, according to the first road element association relationship between the first sub road data and the benchmark road data; transforming the first sub road data according to the first transformation relationship to obtain transformed first sub road data; determining a second transformation relationship from the second sub road data to the benchmark road data, according to the second road element association relationship between the second sub road data and the benchmark road data; transforming the second sub road data according to the second transformation relationship to obtain transformed second sub road data; and fusing the transformed first sub road data, the transformed second sub road data, and the benchmark road data to update the benchmark road data.

6 . The method according to claim 1 , wherein determining the benchmark road data and the at least one road data to be fused in the target road area comprises:
determining a plurality of road data in the target road area; determining one of the plurality of road data as the benchmark road data; and determining road data in the plurality of road data other than the benchmark road data as the road data to be fused.

7 . The method according to claim 6 , wherein determining the plurality of road data in the t

Assignees

• Apollo Intelligent Driving Technology Beijing Co Ltd

Inventors

• Zhou Yao
• Zhang Hantian
• Peng Liang
• Wan Guowei

Classifications

• G06T2207/30241

  Trajectory · CPC title

• G06T7/10

  Segmentation; Edge detection (motion-based segmentation G06T7/215) · CPC title

• G06T11/60

  Creating or editing images; Combining images with text · CPC title

• G01C21/3863

  Structures of map data · CPC title

• G01C21/3859

  Differential updating map data · CPC title