Automatic driving environmental perception system based on vehicle-road coordination

Abstract

The invention discloses an automatic driving environmental perception system based on vehicle-road coordination. The system includes a roadside sensor, a roadside processing unit and a roadside communication unit which are arranged on a road segment divided according to a set distance. The roadside sensor includes a camera, a laser radar, and a microwave radar, and is used for acquiring road information of a road. The roadside processing unit is used for receiving the road information acquired by the roadside sensor, recognizing and processing the road information, then converting the road information into a united world coordinate system, fusing and marking the converted data in a high-precision map, splicing the high-precision maps of adjacent road sections to form semantic road environment information, and transmitting the semantic road environment information to the roadside communication unit. The roadside communication unit is used for sending out the semantic road environment information by using wireless communication technology. The automatic driving environmental perception system, by using the roadside sensor, updates the high-precision map in real time according to theacquired road dynamic characteristic data on the basis of the original high-precision map, so as to provide a global environmental perception capability for the safe driving of an autonomous vehicle.

Description

technical field [0001] The invention belongs to the technical field of intelligent networked transportation, and in particular relates to an automatic driving environment perception system based on vehicle-road coordination. Background technique [0002] As the most cutting-edge technology in the field of intelligent transportation, autonomous driving technology includes vehicle environment perception, path planning and decision-making, and vehicle motion control. At this stage, with the development of artificial intelligence and deep learning processing chips, the three autonomous driving technologies have all developed rapidly. However, in order to realize the vehicle’s comprehensive perception of the surrounding environment, each self-driving vehicle needs to be equipped with complex, expensive, and various sensor equipment, resulting in high vehicle costs. According to the evolution of the autonomous route, the commercialization of self-driving cars The process is slow....

AI Technical Summary

Problems solved by technology

[0004] (1) The vehicle perception range is greatly affected by the performance of the sensor. The current maximum perception distance is about 200 meters, and the perception range for high-speed vehicles is even smaller;

[0005] (2) The perception range is easily affected by obstacles, and it is difficult to achieve a comprehensive perception of the surrounding environment. The current autonomous vehicle perception technology cannot realize the situation ahead by crossing obstacles, and cannot solve the problem of driving safety under the condition of visual obstruction;

[0006] (3) The independent environment perception system of each self-driving vehicle brings challenges to the autonomous driving formation driving. It is difficult to integrate the sensor data of all vehicles in the fleet when the autonomous driving vehicles are driving in formation, resulting in poor robustness of the fleet system and easy disintegration;

[0007] (4) The current high-precision maps have problems such as low update frequency and high acquisition and production costs, which are difficult to meet the actual needs of autonomous vehicles

Images