MODEL PREDICTIVE CONTROL FOR AUTONOMOUS VEHICLE NAVIGATION

Abstract

Model Predictive Control (MPC) has emerged as a powerful control strategy for autonomous vehicle navigation due to its ability to handle multi-variable systems and explicit constraint handling. This paper explores the application of MPC in path planning and trajectory tracking for autonomous ground vehicles. By leveraging a predictive model of vehicle dynamics, MPC enables real-time optimization of control inputs to achieve safe, smooth, and efficient navigation in dynamic environments. The controller accounts for vehicle kinematics, road boundaries, obstacle avoidance, and dynamic constraints, making it particularly suitable for complex driving scenarios such as urban environments. Simulation results and real-world experiments demonstrate the effectiveness of MPC in maintaining lane discipline, executing turns, and avoiding obstacles, while ensuring stability and passenger comfort. Keywords: Model Predictive Control (MPC), Autonomous Vehicles, Trajectory Tracking, Path Planning, Obstacle Avoidance, Vehicle Dynamics, Real-Time Control, Urban Navigation.