Simulation test method and device of automatic driving system and storage medium

Abstract

The invention discloses a simulation test method and device of an automatic driving system and a storage medium, and is applied to the field of automatic driving. The method comprises the steps of obtaining a target test scene in the automatic driving system; adjusting the first control algorithm based on a preset decision to obtain a second control algorithm; further, performing at least two simulation tests in the target test scene based on a second control algorithm to obtain first test data and second test data; and comparing the first test data with the second test data to obtain a test result for indicating the consistency of the automatic driving system. Therefore, the stability consistency test process is realized, the variable parameters in the control algorithm are constrained, the interference of the control algorithm on the automatic driving system test is removed, and the accuracy of the test result is improved.

Description

technical field [0001] The present application relates to the field of computer technology, and in particular to a simulation test method, device and storage medium of an automatic driving system. Background technique [0002] Autonomous driving technology includes high-precision maps, environmental perception, behavior decision-making, path planning, motion control and other technologies, and has broad application prospects. In the process of testing autonomous driving, the reliability of the simulation results of autonomous driving is closely related to whether the simulation environment is similar to the real environment. [0003] In the process of automatic driving simulation, the simulation consistency is an important performance index, that is, in the automatic driving simulation system, after running the same simulation test scene many times, whether the data and timing are consistent and repeatable sex. [0004] However, since the process of automatic driving simul...

AI Technical Summary

Problems solved by technology

[0004] However, since the process of automatic driving simulation generally uses deep learning models for route planning, direction adjustment, etc., it is easy to generate algorithm deviations and affect the accuracy of the consistency test results in the automatic driving system.

Images