Abstract
Electric automated vehicles are zero-emission, energy-saving, and environmentally friendly vehicles, and testing and verification is an important means to ensure their safety. Because of the scarcity of dangerous scenarios in natural driving roads, it is required to conduct accelerated tests and evaluations for electric automated vehicles. According to the scenario data of the natural road in cut-in conditions, we used the kernel density estimation method to calculate the probability distribution of the scenario parameters. Additionally, we used the Metropolis–Hastings algorithm to sample based on the probability distribution of the parameters, and the Euclidean distance was combined with the paired combination to accelerate the simulation test process. The critical scenarios were screened out by the safety indicator, and the feature distribution of the critical scenario parameters was analyzed based on the Euclidean distance clustering method, so as to design importance sampling parameters and carry out importance sampling. The study obtained the distribution characteristics of critical scenario parameters under cut-in conditions and found that the importance sampling method can accelerate the test under the condition of ensuring that the relative error is small, and the improved accelerated simulation method makes the overall calculation amount smaller.
Highlights
Accepted: 13 November 2021In recent years, electric automated vehicle technology has developed rapidly
If the Euclidean distance between an inner loop test case and an ordinary test case was less than the threshold of 0.1, the inner collision condition pre-collision condition dangerous condition safe condition
Euclidean distance and pairwise combination methods were used to improve the computational amount of simulation methods to improve test efficiency
Summary
Accepted: 13 November 2021In recent years, electric automated vehicle technology has developed rapidly. It is an important measure to ensure the safety of electric automated vehicles by functional verification and certification. Due to the complexity and diversity of the natural traffic environments and driving tasks, it has brought great challenges to the testing, verification, and certification for electric automated vehicles. Most of the scenarios encountered during vehicle driving are safe, natural scenarios, with few dangerous scenarios and a low probability of accidents. It requires hundreds of millions of kilometers of driving mileage to test and verify the safety of automated vehicles on natural driving roads, which have a long test period and a high test cost. How to reduce the workload of test and verification and conduct efficient test verification and certification of the safety of automated vehicles has become an important issue that needs to be resolved
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
