Robotic positioning method based on line segment matching within a window

By combining a single-point ranging sensor and a gyroscope, and utilizing line segment matching technology, the high storage space and computational complexity issues in existing SLAM technologies are solved, achieving efficient, real-time robot localization with strong adaptability.

CN117289689BActive Publication Date: 2026-06-09AMICRO SEMICONDUCTOR CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
AMICRO SEMICONDUCTOR CO LTD
Filing Date
2022-06-20
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

Existing SLAM technology relies on rotating LiDAR or vision-assisted positioning, resulting in large storage space requirements, high computational complexity, high cost, poor real-time positioning performance, and easy positioning failure due to error accumulation.

Method used

By employing a single-point ranging sensor and a gyroscope, point cloud data is collected through robot rotation and a line feature sub-map is constructed. Local area localization is achieved by utilizing line segment matching, thereby reducing computational complexity and minimizing error accumulation.

Benefits of technology

It achieves efficient, real-time, and robust localization within a local area, reducing computational complexity and storage requirements. It is highly adaptable and requires no additional sensor modifications.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117289689B_ABST
    Figure CN117289689B_ABST
Patent Text Reader

Abstract

The application discloses a robot positioning method based on line segment matching in a window, comprising the following steps: a robot collects point cloud data through a single-point ranging sensor and constructs a current window; the robot rotates at different rotation center positions in the current window in sequence, and fits and processes the collected point cloud data at each rotation center position to obtain a corresponding group of fitted line segments, and then respectively forms a line feature subgraph from each group of fitted line segments; the robot controls the fitted line segments in each line feature subgraph to perform similarity matching with the fitted line segments in other line feature subgraphs; when the matching success rate of the fitted line segments in two line feature subgraphs reaches a first preset success rate, an error coordinate offset is obtained; when the matching success rate between the line feature subgraphs in the current window reaches a second preset success rate, an average value of all error coordinate offsets is processed to obtain a positioning coordinate compensation, and then the positioning coordinate compensation is used to correct the current position coordinate of the robot.
Need to check novelty before this filing date? Find Prior Art