Real-time road segmentation method based on 3D (three-dimensional) laser radar

Abstract

The invention provides a real-time road segmentation method based on a 3D (three-dimensional) laser radar. The real-time road segmentation method based on the 3D laser radar includes the steps of: (1)conducting acquisition and processing of raw 3D lidar data; (2) extracting a single line and drawing an azimuth sequence-distance map; (3) using an improved fuzzy line segment method to map a line segment map and identifying road surfaces and obstacles; (4) removing obstacles, identifying the road surface and outputting final passable areas. The real-time road segmentation method based on the 3Dlaser radar has the advantages of well adapting to the identification of slowly changing road surfaces with inclination angles via the improved fuzzy line segment method, shortening subsequent line segment processing cycles with the improving traditional fuzzy line segment method to handle segment segmentation problem of of the slowing changing road surfaces with the inclination angles, and providing high accurate and stable extraction results.

Description

technical field [0001] The invention relates to the field of unmanned driving technology, in particular to a real-time road segmentation method based on three-dimensional laser radar. Background technique [0002] An unmanned vehicle is a vehicle that senses the road environment through an on-board sensor system, thereby automatically planning a driving route and controlling the vehicle to reach a predetermined target. Use on-board sensors to perceive the surrounding environment of the vehicle, and control the steering and speed of the vehicle according to the road, vehicle position and obstacle information obtained by the perception, so that the vehicle can drive safely and reliably on the road. Driverless cars often use lidar and cameras to perceive the road. 3D lidar can accurately obtain 3D environmental information in real time through non-contact scanning, with little environmental interference and high precision. [0003] The patent No. CN201210000399.4 is an invent...

AI Technical Summary

Problems solved by technology

[0004] The above inventions are all based on extracting effective line segments to identify characteristic objects or roads, but the number and effectiveness of effective line segments are accidental, and the effectiveness and number of line segments affect the real-time and accuracy of data processing to a large extent. In the extraction of traversable roads for unmanned vehicles, it is also necessary to minimize the number of processing line segments under the premise of ensuring the validity of the line segments. For example, in urban roads, the road surface is not completely flat, and the line segments are segmented on slightly uneven roads. The phenomenon is more serious, the number of line segments is large, resulting in waste of data processing

Images