一种激光雷达自定位和三维环境重构方法及装置
By combining the CTRV model and the NDT algorithm, point cloud distortion correction and registration are performed. An adaptive resolution strategy is adopted to solve the problem of point cloud map density over time, and high-precision 3D environment reconstruction is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI UNIV
- Filing Date
- 2023-07-21
- Publication Date
- 2026-07-17
AI Technical Summary
In existing technologies, point cloud maps become increasingly dense over time, generating a large number of redundant points, and traditional methods struggle to effectively distinguish objects with different features and achieve map fusion.
Distortion correction is performed using a constant turning rate and velocity model (CTRV), point cloud registration is performed using the NDT algorithm, point clouds are merged using an adaptive resolution strategy, the point cloud fusion density is optimized through the adaptive resolution strategy and curvature calculation, redundant points are removed, and a high-precision global map is formed.
It effectively controls the size of point cloud maps, improves map building accuracy, reduces redundant points, and enhances the accuracy and efficiency of map building, making it suitable for both structured and unstructured environments.
Smart Images
Figure CN116883596B_ABST