Risk Evaluation during the Takeover of Automated Vehicles Based on Improved Driving Safety Field

A comprehensive and accurate evaluation of takeover risk is key to improving the safety of automated vehicles. The research on takeover mainly focuses on the factors affecting takeover performance, and rarely involves the quantification of the risk of automated vehicles executing a takeover. This study considers the influence of physical object shape on field distribution, in order to improve the existing driving safety field. On this basis, the safety index of a takeover was established, and a risk evaluation method of takeover considering multiple risk sources was proposed. The experimental results show that the improved field strength distribution is more consistent with the actual situation, and it solves the problem of overestimating and underestimating risk. The proposed takeover risk evaluation method considering multiple risk sources has good rationality.