Open AccessAutonomous‐driving vehicle test technology based on virtual reality
Author(s) -
Yao Shouwen,
Zhang Jiahao,
Hu Ziran,
Wang Yu,
Zhou Xilin
Publication year - 2018
Publication title -
the journal of engineering
Language(s) - English
Resource type - Journals
ISSN - 2051-3305
DOI - 10.1049/joe.2018.8303
Subject(s) - virtual reality , computer science , simulation , driving simulator , controller (irrigation) , automotive engineering , test (biology) , hardware in the loop simulation , driving test , scenario testing , engineering , human–computer interaction , artificial intelligence , biology , paleontology , logistic regression , variety (cybernetics) , machine learning , agronomy
In order to mitigate risks from road tests for autonomous‐driving vehicles, reduce costs and accelerate development, a virtual reality (VR)‐based test platform for autonomous‐driving vehicles was built combined with the AirSim system and the UE4 engine by establishing a model library which contains the vehicle dynamics model, sensor models and traffic environment model. The controller‐in‐the‐loop simulation method was implemented to complete the simulation test for autonomous vehicles under different driving conditions and the simulation results were used to optimise the autonomous‐driving control system. The actual autonomous driving road test can now be done in an immersive VR simulation environment where autonomous‐driving road‐testing is done safely and cost‐effectively. This plays a significant role in the future development of autonomous‐driving vehicles.