A Hierarchical Framework of Decision Making and Trajectory Tracking Control for Autonomous Vehicles

Most of the existing research in field autonomous vehicles (AVs) addresses decision making, planning and control as separate factors which may affect AV performance complex driving environments. A hierarchical framework is proposed this paper to address problem mentioned above environments with multiple lanes surrounding vehicles. Firstly, high-level making implemented by a finite-state machine (FSM), according relative relationship between ego vehicle (EV) After made, cluster quintic polynomial equations established generate path connecting initial position candidate target positions, traffic states EV vehicle. The optimal selected from cluster, based on quadratic programming (QP) framework. Then, speed profile generated, longitudinal displacement–time graph (S–T graph), considering motion state constraints collision avoidance. smoothed created through another QP formulation, space dynamic-programming (DP) method. Finally, planned are combined sent lower level module, consists linear regulator (LQR) for lateral trajectory tracking control, double PID controller control. was validated co-simulation environment using PreScan, MATLAB/Simulink CarSim, results demonstrate that capable addressing most ordinary scenarios structured road, reasonable decisions controlling abilities.