Map construction and maintenance method and device based on visual landmarks and storage medium

By introducing visual landmark information and optimizing objective function constraints, the problems of positioning drift and map misalignment in the laser SLAM algorithm are solved, enabling long-term maintenance and accurate construction of high-quality maps.

CN117570968BActive Publication Date: 2026-06-26TONGJI UNIV
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
TONGJI UNIV
Filing Date
2023-12-07
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

Existing laser SLAM algorithms suffer from problems such as localization drift, map misalignment, and map loss in mobile robot localization and map building, and map quality is difficult to guarantee, especially during long-term maintenance.

Method used

By introducing visual landmark information and increasing the constraints of the graph optimization objective function through the pose relationship between visual landmarks and laser frames, and by combining visual landmark pose updates and filtering dynamic obstacle data, redundant subgraphs are pruned to optimize the map construction process.

Benefits of technology

It improves the positioning accuracy and map quality of map construction, reduces the complexity of map construction and optimization, and ensures the long-term maintenance efficiency and accuracy of maps.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117570968B_ABST
    Figure CN117570968B_ABST
Patent Text Reader

Abstract

The present application relates to the field of mobile robot autonomous mapping and positioning, in particular to a kind of map construction and maintenance method, device and storage medium based on visual road mark.The method obtains laser frame pose, subgraph pose and first pose relationship by using front-end scan matching and loop detection, obtains first constraint condition;Obtain visual road mark pose and second pose relationship by using environment detection, obtain second constraint condition;Solve the graph optimization objective function constructed according to the first constraint condition and the second constraint condition, obtain the first map;Filter dynamic obstacle data in the first map, prune redundant subgraph, and update and solve the graph optimization objective function in combination with the corresponding visual road mark pose, obtain the second map.Compared with prior art, the present application has the advantages of effectively improving the positioning accuracy and map quality of map construction, while realizing long-term maintenance of high-quality map.
Need to check novelty before this filing date? Find Prior Art