Study of the Hazard Perception Model for Automated Driving Systems

Automated and human-driven vehicles will coexist for a long time. It would be helpful to improve user experience of automated vehicles by considering drivers' psychological model of hazard perception. This work attempts to build a hazard perception model of a typical traffic scenario for automated driving systems. Seventeen drivers were recruited as participants for the driving simulation experiment to investigate the effects of different road conditions on drivers' subjective assessment of danger level and risk acceptance. A nonlinear regression model of hazard perception was built based on the experimental results. A case study has shown that the model can effectively reflect the quantitative relationship between drivers' perceived danger level and the relevant road conditions. It will provide theoretical basis for the development of future automated driving systems for users with different risk preferences.