Method, system and device for detecting road target and distance of automatic driving vehicle

Abstract

The invention belongs to the technical field of automatic driving, particularly relates to a method, a system and a device for detecting a road target and a distance of an automatic driving vehicle, and aims to solve the problems of low speed and low precision when the road target and the distance of the automatic driving vehicle are detected in the prior art. The method comprises the following steps: acquiring a road target image and point cloud data of an automatic driving vehicle, and respectively preprocessing the road target image and the point cloud data; obtaining category label information and 2D bounding box information of the target through a target detection network; performing fusion processing on the road target image and the point cloud data to obtain point cloud with image label information in a 2D bounding box; and obtaining a target point cloud through target detection, and taking the distance of the mass center of the target point cloud as the distance of the currenttarget. The method can calculate the actual distance of the target, is not liable to be affected by complex weather and geographical environment, just needs to process the point cloud data of the 2D detection frame, is high in speed, and is high in precision.

Description

Technical field [0001] The present invention belongs to the field of automatic driving technology, and specifically relates to a method, system, and device for detecting road targets and distances of automatic driving vehicles. Background technique [0002] Self-driving vehicles can also be called wheeled automatic robots, which mainly rely on the on-board sensor system to perceive the road environment, automatically plan the driving route and control the vehicle to reach the predetermined goal. Autonomous vehicles are highly developed products that integrate many technologies such as automatic control, visual computing, and architecture. Environment perception, precise positioning, path planning, and wire control execution are the core technologies of autonomous driving. Among them, environment perception directly determines whether the subsequent roadside planning decisions are accurate, so it is very important. In environment perception, if information such as the category an...

AI Technical Summary

Problems solved by technology

[0009] In order to solve the above problems in the prior art, that is, the speed is not fast enough and the accuracy is not high enough when the prior art realizes the road target and distance detection of the automatic driving vehicle, the present invention provides a road target and distance detection method for the automatic driving vehicle. The detection method, the target and distance detection method includes:

Images