Research on screening methods of autonomous driving dangerous test cases based on Carmaker

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Autonomous driving vehicle is the mainstream trend of the development of the automotive industry at home and abroad. In the process of its research and development, many safety problems are frequently exposed. Therefore, the safety test of autonomous driving vehicles is particularly important. Autonomous driving vehicle testing can be carried out in three aspects: virtual simulation test, proving ground test and real road test. Among them, proving ground test, as a kind of real vehicle test, mainly tests typical dangerous scenarios. Based on the simulation software Carmaker, a method of automatic generation of simulation test cases is proposed in this paper. The method defines different parameter variables and parameter boundary, realizes the automatic combination of parameters through the editing and running of the code, and improves the efficiency of simulation test and the coverage of test cases. Through the high-efficiency and high-coverage hardware-in-the-loop simulation test, the dangerous test cases of typical scenarios are screened out for the proving ground test, so as to make the proving ground test more targeted and improve the efficiency of the proving ground test. Meanwhile, combination of virtual simulation and proving ground in actual test is able to ensure simulation result effectiveness.

Original languageEnglish
Title of host publicationInternational Conference on Smart Transportation and City Engineering 2021
EditorsZhengliang Li, Fengjie Cen
PublisherSPIE
ISBN (Electronic)9781510649750
DOIs
Publication statusPublished - 2021
Externally publishedYes
EventInternational Conference on Smart Transportation and City Engineering 2021 - Chongqing, China
Duration: 26 Oct 202128 Oct 2021

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume12050
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

ConferenceInternational Conference on Smart Transportation and City Engineering 2021
Country/TerritoryChina
CityChongqing
Period26/10/2128/10/21

Keywords

  • Autonomous driving
  • Carmaker
  • Dangerous test cases
  • Proving ground test
  • Simulation test

Fingerprint

Dive into the research topics of 'Research on screening methods of autonomous driving dangerous test cases based on Carmaker'. Together they form a unique fingerprint.

Cite this