Personalized Autonomous Vehicle Control for Typical Driving Scenarios

In a real traffic environment, different drivers demonstrate different driving behaviors, and a single driving mode of an autonomous vehicle is not able to meet different driving requirements. Especially, for high-level autonomous vehicles, if the autonomous vehicle does not match the human driver's expectations, the acceptance of the autonomous vehicle system will be reduced, and sometimes may even lead to accidents. This paper fully considers the personalized driving characteristics for autonomous vehicle control. Firstly, with the control mode switching algorithm, autonomous vehicles realize the switch of control modes with respect to different typical driving scenarios. Secondly, based on chance constrained programming, this study introduces the driver into control constraints, which can reflect different driving personality characteristics. Finally, the proposed algorithms are tested in the hardware in the loop (HIL) experiment. The results show that the proposed personalized algorithm and control mode switching method can not only improve the comfort of the vehicle, but also ensure the safety of the driver.