Road running and virtual test parallel mapping experiment method of intelligent automobiles

Abstract

The invention relates to a road running and virtual test parallel mapping experiment method of intelligent automobiles. The objectives of the invention are to improve efficiency and safety of intelligent automobile test. The method comprises steps of firstly acquiring track data of an automobile driven by a driver, interception data of an automatic driving system and planed track data (wherein theintelligent automobile fails to execute the planed track at this moment); then comparing an actual track and the planned track, and searching whether there is a remarkably different scene between thetwo tracks; and then, through the interception information of the intelligent automobile, reconstructing the scene, carrying scene replaying and simulating prediction, calculating safety of the two tracks and evaluating the safety of the tracks. According to the invention, there is no need to use an automatic driving system to control the automobile; and as long as the intelligent automobile runson a road (i.e., a driver is operating the automobile), the test and the experiment can be carried out. Thus, test efficiency can be improved and safety of the test can be ensured.

Description

technical field [0001] The invention belongs to the technical field of smart car testing, and more specifically, the invention relates to a process of searching for a scene where there is a difference between the planned route of the smart car and the driving route of the driver when the driver drives the smart car autonomously on an actual road , and build the same scene in the virtual environment, and carry out scene playback and simulation test, so as to realize the system of testing and evaluating the driving safety of smart cars. Background technique [0002] With the development of science and technology, smart cars based on autonomous driving technology have shown great development potential in improving traffic safety and preventing traffic congestion, which is the direction of the development of the automotive industry and traffic operations. [0003] The automatic driving system of a smart car consists of three modules: environment perception, planning decision-mak...

AI Technical Summary

Problems solved by technology

In addition, public road testing has high traffic safety risks and is difficult to carry out on a large scale due to factors such as laws and policies

The proving ground test is often a functional test of a single vehicle under a simple traffic scene in a fixed scene, and it is difficult to fully and truly reproduce the complex environment that a smart car may encounter on the road.

Virtual testing can improve the efficiency of testing and ensure the safety of testing, but it cannot guarantee the authenticity of testing scenarios

Images