Path planning in a multi-obstacle environment with consideration of driver’s individual preferences

Diverse drivers have various driving habits and driving preferences, the function of personalized driving is one of the most important features for autonomous vehicles (AVs). In this paper, a tunable-driver style factor is proposed to describe the individual aggressive preferences of different drivers. Based on this, a hierarchical human-like lane-changing path planning framework is proposed. Firstly, an alternate trajectory set (ATS) method is proposed to obtain a set of trajectories that are translational to the road centerline. The artificial potential field (APF) method is deployed to select the optimal safe trajectories subject to the environmental constraints. Then, the B-spline polynomial curve is utilized to generate safe and smooth lane-changing trajectories with consideration of the driver style factor. Finally, a linear time-varying (LTV) model predictive control (MPC) method is adopted to track the planned trajectories. Simulation results show that the proposed framework can generate safe and personalized trajectories for different drivers in complex environments with multi-obstacles.