Semantic high-precision map construction and positioning method based on point-line feature fusion laser

Abstract

The invention discloses a semantic high-precision map construction method based on dot-line feature fusion laser, which comprises the following steps: 1) performing semantic segmentation and feature extraction on a visual image acquired by a camera to obtain a visual image containing semantic categories and dot-line features, and then obtaining a foreground and a background of a moving target; 2)projecting the laser three-dimensional point cloud acquired by the laser radar onto a visual image plane, fitting to obtain a depth map, and endowing the laser three-dimensional point cloud with semantic categories and dot line features; 3) performing super-pixel segmentation on the moving target, calculating the distance between super-pixel blocks, constructing a graph model, performing image segmentation, and accurately extracting the boundary of the moving target; and 4) removing the visual feature points and the laser three-dimensional points belonging to the moving target to construct a static high-precision semantic three-dimensional map. The invention further discloses athe positioning method of the semantic high-precision map constructed by adopting the method, and the mapping is more accurate and reliable by accurately removing the moving target.

Description

technical field [0001] The invention relates to a semantic high-precision map, in particular to a semantic high-precision map construction and positioning method based on point-line feature fusion laser. Background technique [0002] High-precision positioning is the basis for realizing the automatic driving of unmanned vehicles, and provides a basic guarantee for path planning and motion control of unmanned vehicles. To achieve high-precision positioning of unmanned vehicles, high-precision maps are an indispensable part. [0003] High-precision maps usually contain high-precision point cloud maps and semantic information such as lane lines, signal lights, and roadsides. Semantic elements such as lane lines, curbs, traffic signs, and signal lights of traditional high-precision 3D maps are manually edited and added on the constructed 3D point cloud map, which requires a lot of manpower and material resources. At the same time, because the surrounding environment is always ...

AI Technical Summary

Problems solved by technology

Compared with traditional GPS-based positioning, high-precision map-based matching positioning is not affected by GPS signals, and can still operate in tunnels, elevated environments, etc.; however, laser matching-based positioning methods rely heavily on 3D high-precision maps. Therefore, when the scene changes, the high-precision map is prone to mis-matching or unsuccessful matching, and the map needs to be updated continuously, resulting in high operating costs.

Images