A 3D point cloud map fusion method and system for a substation inspection robot

Abstract

A 3D point cloud map fusion method and system for a substation inspection robot of the present invention includes the following steps: before the inspection robot scans the image, the scene to be inspected is divided into local areas, and the scanning path is planned for each local area. ; Control the inspection robot to scan multiple local areas, each local area has a closed loop and a loop is detected, and the adjacent areas fused have a common part; sequentially import the 3D point cloud map to be registered and the reference 3D point cloud map, Manually frame the same areas in the two maps; perform rough-to-fine matching on the same areas in the above-mentioned frame-selected two maps through a pre-set fusion algorithm, and obtain a fusion map. The 3D point cloud map fusion method of the substation inspection robot of the present invention can avoid the risk of global optimization failure, realize the high-precision fusion of multi-local 3D point cloud maps of substation scenes of more than 100,000 square meters, and satisfy the positioning and navigation of the inspection robot.

Description

technical field [0001] The invention relates to the technical field of intelligent robots, in particular to a 3D point cloud map fusion method and system for a substation inspection robot. Background technique [0002] SLAM is one of the core functions of wheeled intelligent inspection robots, and 3D laser SLAM is a research hotspot in this field in recent years. It is also a trend in recent years to replace 2D laser SLAM. Substation intelligent inspection robots are one of its application carriers. 500KV and above substations have typical characteristics such as large scale, complex inspection routes, open space and large observation noise. The larger the environment, the longer the time required for data collection and map construction, and the larger the cumulative error; the more complex the inspection route, the higher the accuracy requirements for closed-loop detection; the sparser features and the larger observation noise, the larger the matching error . The above f...

AI Technical Summary

Problems solved by technology

[0007] [4] CN110930495A, ICP point cloud map fusion method, system, device and storage medium based on multi-UAV cooperation. This method mainly uses SAC-IA to estimate the corresponding point pairs, and screens the point pairs, and then uses ICP for matching. When the transformation of the two maps is large, it is easy to fail the fusion

Images