Underwater positioning and mapping method based on improved key frame decision and attitude optimization

By improving the keyframe decision-making and attitude optimization methods, combining IMU observation information and sliding mode filtering algorithm, the pose drift and positioning error problems of SLAM system in dynamic underwater environment and vigorous movement are solved, and the accuracy and robustness of the system are improved.

CN119687904AActive Publication Date: 2025-03-25HOHAI UNIV
11 Cites 1 Cited by

Patent Information

Application Number
CN202510205752.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-02-25
Publication Date
2025-03-25
Estimated Expiration
2045-02-25

AI Technical Summary

Technical Problem

When the SLAM system is operating underwater for a long time, it may encounter dynamic environment, violent AUV movement or rapid bending movement at a large perspective, resulting in pose drift and positioning errors.

Method used

By improving the keyframe decision-making method, combining IMU observation information to judge the dynamic environment and vigorous movement, select the main and secondary keyframes and establish double measurement constraints, use the sliding mode filtering algorithm to optimize the camera position, restore the local trajectory and generate new map points.

Benefits of technology

It improves the accuracy and robustness of the visual inertial SLAM system, reduces pose drift and positioning errors, and enhances the system's perception of motion state and keyframe selection accuracy.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses an underwater positioning and mapping method based on improved key frame decision and attitude optimization. The method comprises the following steps: acquiring data of a sensor configured by the AUV; the method comprises the following steps of: processing obtained data of a sensor, initializing system parameters, tracking a local map, judging whether a current environment is a dynamic environment or not and whether an AUV (Autonomous Underwater Vehicle) does strenuous motion or not according to observation information of an IMU (Inertial Measurement Unit) on the basis of ORB-SLAM3 key frame selection, and continuing to select key frames; optimizing the attitude of the selected key frame, performing attitude correction by using double constraints, recovering a local track under the large-view-angle rapid bending motion, and generating a new map point; carrying out closed-loop detection and correction; and generating a global track and a map. According to the method, the problem of pose drift caused by the fact that the autonomous underwater vehicle possibly encounters a dynamic environment or the autonomous underwater vehicle moves violently is solved, and meanwhile, the accuracy and robustness of a visual inertia synchronous positioning and mapping system are greatly improved.
Need to check novelty before this filing date? Find Prior Art