Point cloud descriptor construction method and device and closed-loop detection method and device

Abstract

The invention provides a point cloud descriptor construction method and device, a closed-loop detection method and device, electronic equipment and a storage medium, the point cloud descriptor construction method is applied to space positioning, the point cloud descriptor construction method comprises the steps of acquiring a point cloud which is point cloud data acquired through a laser detector;extracting a plurality of line segments from the point cloud, wherein every two of the plurality of line segments form a line segment pair; determining a geometric invariant feature of each line segment pair; and constructing a point cloud descriptor according to the geometric invariant features. According to the method, the line segment pairs are extracted from the laser point cloud, the geometric invariant features of the line segment pairs are determined, and the point cloud descriptor is constructed based on the geometric invariant features, so that the point cloud descriptor has rotationinvariance, and the problem that a closed-loop detection algorithm cannot work stably due to the fact that the point cloud descriptor is constructed based on feature points at present is solved.

Description

technical field [0001] The present application relates to the technical field of navigation equipment positioning, in particular, to a point cloud descriptor construction method and device, a closed-loop detection method and device, electronic equipment and storage media. Background technique [0002] At present, when navigation devices (for example, mobile robots, aircraft, etc.) use single-line lasers to perform track deduction in the environment, the deduced trajectory is inaccurate due to the existence of accumulated errors. In the prior art, a closed-loop detection (LoopClosure) technology can be used to identify historical scenes during the movement of the navigation device, thereby eliminating accumulated errors. Therefore, loop closure detection is more important for map construction and positioning of navigation equipment. [0003] Generally, since the image information of the visual sensor is more abundant than the point cloud data obtained by the single-line lase...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the embodiment of the present application is to provide a point cloud descriptor construction method, a point cloud descriptor construction device, a closed-loop detection method, a closed-loop detection device, electronic equipment and a storage medium, so as to improve the construction of points based on feature points in the prior art. The cloud descriptor makes the closed-loop detection algorithm unable to work stably, which in turn makes the recognition result inaccurate when identifying historical scenes in the process of navigation equipment movement based on closed-loop detection

Images