A Vehicle-in-the-Loop Simulation Test Based Digital-Twin for Intelligent Vehicles

Evaluating the performance of intelligent vehicle (IV) accurately and efficiently is crucial to the development and deployment of IV. The previous test methods for IV performance evaluation mainly include use-case-driven and scenario-driven test methods. Scenario-driven test method is widely adopted because of good performance. According to the test system, IV test tools mainly include virtual simulation testing, closed field testing, and open road testing, the three test tools have different focuses. Existing test methods and tools are either expensive or poor in test accuracy, they also have limitations such as difficult reuse of scenarios, Large test space requirements, etc. Digital-twin technology provide the possibility for the integration of multiple test tools. This paper proposes a digital-twin technology-based vehicle-in-the-loop (VIL) test method, a prototype system also be given. The proposed prototype system includes a virtual scenario generation subsystem, a road simulation platform subsystem, and a monitoring & evaluation subsystem. The virtual scene generation subsystem can generate scenarios and transmit to the IV's perception devices; the road simulation platform shelf system is used to simulate the road conditions during the IV driving, and the monitoring & evaluation subsystem is used to monitor and evaluate the performance of the IV in real time. The proposed VIL-test prototype system can efficiently and repeatedly test and evaluate the performance of IV.