Automatic driving vehicle test method and device, controller and medium

Abstract

The invention relates to an automatic driving vehicle testing method and device, a controller and a medium. The method comprises: acquiring the longitudinal distance between an automatic driving vehicle to be tested and a cooperative testing vehicle and the speed of the cooperative testing vehicle; determining a target acceleration of the collaborative test vehicle based on the longitudinal distance, the vehicle speed of the collaborative test vehicle and a preset expected speed and an expected longitudinal distance of the collaborative test vehicle; and adjusting the running state of the collaborative test vehicle according to the target acceleration. According to the invention, a test scene is provided for accurate data test of the automatic driving vehicle, the accuracy of a test resultis improved, and the test cost is reduced.

Description

technical field [0001] The present invention relates to the technical field of automatic driving, in particular to an automatic driving vehicle testing method, device, controller and medium. Background technique [0002] Before the road test of self-driving vehicles, it is necessary to conduct sufficient real vehicle tests in closed venues in combination with the actual conditions in various places. Existing tests for self-driving vehicles are mainly to set up scenes in the closed test area of ​​self-driving, such as the test of 2021 CNCAP standard scene, mostly use control robots to control the vehicles in the test scene, and also need to attach a set of computing units to the vehicle And the actuator to realize the control of the vehicle, adjust the steering wheel and the brake accelerator pedal through the mechanical arm, not directly control the vehicle itself, the cost is very high. [0003] In addition, in the process of setting up the scene, there is usually no preci...

AI Technical Summary

Problems solved by technology

Existing tests for self-driving vehicles are mainly to set up scenes in the closed test area of ​​self-driving, such as the test of 2021 CNCAP standard scene, mostly use control robots to control the vehicles in the test scene, and also need to attach a set of computing units to the vehicle And the actuator to realize the control of the vehicle, adjust the steering wheel and the brake accelerator pedal through the mechanical arm, not directly control the vehicle itself, the cost is very high

[0003] In addition, in the process of setting up the scene, there is usually no precise scene design, and only focus on whether the autonomous driving vehicle under test can complete the braking, avoidance actions, etc., and have not built a scene that can quantify the test indicators, so it is impossible to realize different distances in the same scene. , speed and other coverage tests, the test accuracy is low, so it is impossible to make an accurate judgment on whether the functional performance of the autonomous vehicle under test is reasonable and comfortable

Images