Laser radar road edge sensing system of intelligent driving electric sweeper

Abstract

The invention relates to a laser radar road edge sensing system of an intelligent driving electric sweeper, which mainly comprises a selection scheme of the laser radar, an installation position and angle of the laser radar, receiving and processing of asynchronous laser point clouds, establishing and updating of a two-dimensional occupied grid map and segmenting and extracting of road edges, wherein the selection scheme of the laser radar adopts two 16-line laser radar sensors to simultaneously sense a front obstacle and a right side road edge; the installation position and angle of the laserradar effectively avoid a perception blind area and realize the robust detection of the road edge; receiving and processing of the asynchronous laser point clouds ensures the time approximate synchronization of the two radar point clouds by adopting a multithreading asynchronous receiving scheme; according to establishing and updating of the two-dimensional occupied grid map, point cloud information is projected into the grid map by using a vertical projection and altitude difference filtering method, and the probability of the grid map is updated; according to segmenting and extracting of road edges, sample points of the road edges are sampled in a segmented manner, a road edge model is estimated through a least square method, and whether the road edges exist or not is distinguished; andthen accurate and robust reference information for tracking control is provided. Compared with the prior art, the laser radar road edge sensing system of the intelligent driving electric sweeper hasthe advantages of accurate perception, wide application range and the like.

Description

technical field [0001] The invention relates to the technical field of intelligent driving electric sweeping vehicles, in particular to a laser radar roadside sensing system for intelligent driving electric sweeping vehicles, which can provide robust and real-time roadside detection results for sweeping vehicles in the drivable road area, Complete precise curb sweeping tasks. Background technique [0002] Intelligent driving electric sweepers for specific scenarios have higher requirements for the perception of the roadside environment. [0003] The existing road edge detection method based on vision is difficult to guarantee the accuracy of the result due to the monotonous texture features of the road edge; Vision sensors rely too much on the lighting environment, making it difficult to clean around the clock. Among existing laser-based curb detection methods, the method of directly extracting curbs from point cloud data is costly. Therefore, the present invention propos...

AI Technical Summary

Problems solved by technology

[0003] The existing road edge detection method based on vision is difficult to guarantee the accuracy of the result due to the monotonous texture features of the road edge; Vision sensors rely too much on the light environment, making it difficult to clean around the clock

Among existing laser-based curb detection methods, the method of directly extracting curbs from point cloud data is costly

Images