Robot localization methods, mobile robots and storage media
By extracting and matching straight line segments in laser navigation and combining the branch and bound method to search for pose, the problems of low accuracy and computational complexity of laser navigation in complex environments are solved, thus improving navigation accuracy and efficiency.
CN121577036BActive Publication Date: 2026-07-17ZHEJIANG HUARAY TECH CO LTD
2 Cites 0 Cited by
Patent Information
- Application Number
- CN202511686453.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-17
- Publication Date
- 2026-07-17
- Estimated Expiration
- 2045-11-17
AI Technical Summary
Technical Problem
Existing laser navigation technology has low navigation accuracy and high computational complexity in complex environments. In particular, multi-line lasers require processing a large amount of point cloud data, which leads to high performance requirements for robot platforms.
Method used
By acquiring laser point cloud frames and predicting poses, current and reference line segments are extracted, matched, and angles are calculated. The pose is then searched in the laser prior map using the branch and bound method, reducing the number of candidate solutions and lowering computational complexity.
Benefits of technology
This improves the accuracy and efficiency of laser matching, reduces computational complexity, and decreases the performance requirements of the robot platform.
✦ Generated by Eureka AI based on patent content.
Abstract
This application discloses a robot localization method, a mobile robot, and a storage medium. The robot localization method includes: extracting line segments from a laser point cloud frame to obtain multiple current line segments; extracting line segments within the vicinity of the current predicted pose from a laser prior map to obtain multiple reference line segments; matching the multiple current line segments and multiple reference line segments to obtain successfully matched line segment pairs; calculating the angle of the mobile robot based on the line segment pairs; and then searching for the position of the mobile robot in the laser prior map based on the angle using a branch and bound method. By distinguishing between angle and position in pose determination, the geometric information detected by the laser is fully utilized for angle calculation first. When searching for pose in the laser prior map using the branch and bound method, only position search is required, reducing the number of candidate solutions during the matching process, greatly reducing computational complexity, and improving the accuracy and efficiency of laser matching.
Need to check novelty before this filing date? Find Prior Art