Large-scale application of autonomous vehicles requires rigorous testing on actual roads, and the quality and reliability of the test can be guaranteed when dealing with complex traffic risk scenarios. Based on over 500 cases of traffic accidents occurred within Yizhuang Development Zone in Beijing from 2009 to 2017, a risk scenario database was constructed and the accident scenarios were analyzed by deconstructing and reconstructing the variables of accident cases. According to the accident cases, nine field variables composed of main factors such as time, space environment and accident participation were deconstructed, 4,320 effective risk scenarios were reconstructed by artificial neural network, and five typical test scenarios of autonomous vehicles within Yizhuang Development Zone were extracted. The results show that side crash, head-on crash and rear-end crash among all the types account for a large proportions of accident probability for AV testing. The conclusions of the study can offer new ideas for the construction of the risk scenario database, and also provide a valuable reference for the road testing of autonomous vehicles.

CICTP