Autonomous Vehicle Planning and Control Method Based on Interacting Vehicle Trajectory Prediction in Intersection Scenarios

Autonomous driving technology is regarded as an effective approach to improving traffic safety and reducing accidents. However, its performance in complex urban scenarios, particularly at intersections, still requires improvement. To enhance trajectory prediction and tracking control, a planning and control framework based on interacting-vehicle trajectory prediction is proposed. First, a deep learning trajectory prediction model integrating an attention mechanism and driving intention recognition is developed to predict the future trajectories of interacting vehicles with high accuracy. Based on the prediction results, collision-avoidance trajectory planning is performed. Then, a trajectory tracking controller combining compensated explicit model predictive control (C-EMPC) and PID control is designed to achieve both high tracking accuracy and real-time performance. The proposed framework is evaluated in a typical intersection scenario through simulation. Results show that the proposed framework can improve driving safety and efficiency in the tested scenario. Furthermore, hardware-in-the-loop experiments indicate the real-time execution feasibility and effectiveness of the proposed method.