System and method for performing beyond-visual-range perception on complex road conditions by intelligent vehicle

Abstract

The invention discloses a system and a method for performing beyond-visual-range perception on road conditions by an intelligent vehicle; the system comprises the intelligent vehicle and an unmanned aerial vehicle, after the vehicle-mounted unmanned aerial vehicle takes off, the time of the vehicle-mounted unmanned aerial vehicle is synchronized with the time of the intelligent vehicle, and the vehicle-mounted unmanned aerial vehicle starts a following shooting mode, corrects the position in real time, follows the vehicle to move and is kept right above the vehicle; based on the images sent back by the vehicle-mounted unmanned aerial vehicle in real time, an intelligent terminal of the intelligent vehicle operates a target detection algorithm to detect and classify targets, and the targetscomprise lane lines, vehicles, pedestrians, traffic signboards, delimited identification frames and identification IDs; information obtained by a vehicle-mounted visual sensor is supplemented by utilizing an unmanned aerial vehicle image recognition result, specifically, a re-recognition algorithm is carried out on a target detected by an unmanned aerial vehicle image and pedestrian and vehicle images detected by the vehicle-mounted visual sensor, and for shielded traffic lane lines and signboards, the information obtained through identification is directly provided for the intelligent vehicle.

Description

technical field [0001] The invention relates to the field of automatic driving, in particular to a system and method for an intelligent vehicle to sense complex road conditions beyond the horizon. Background technique [0002] Environmental perception in autonomous driving has always been a complex and arduous task. In the actual driving process, the environment is often very complex and diverse, involving traffic sign recognition, traffic light recognition, ground marking recognition, pedestrian recognition, vehicle recognition, road terrain It is difficult to obtain all the information by the sensor of the smart car itself, especially in the case of occlusion. To solve this problem, the existing solution is to use the information captured by the roadside sensors and the mutual fusion of the sensor information of multiple smart cars in the traffic system to form a smart car-linked system that integrates perception and communication. It can also use high-precision maps to he...

AI Technical Summary

Problems solved by technology

[0003] However, the implementation of the above methods requires a high initial investment, such as laying out roadside sensors and drawing high-precision maps in advance, and is not flexible enough and prone to dead ends.

Images