Model Predictive Control based Trajectory Tracking for Autonomous Vehicles

This paper focuses on the development and implementation of a Model Predictive Control (MPC) algorithm for real-time trajectory tracking in a four-wheel autonomous vehicle model. The vehicle features a dual-loop control system: an inner loop governed by an STM32F407 microcontroller for speed and steering angle control, and an outer loop managed by a Jetson Nano to minimize trajectory error and reduce steering oscillations. Equipped with RTK GPS for precise positioning and an IMU for orientation, the vehicle leverages these sensors to ensure accurate and stable navigation. This study details the integration of the MPC algorithm within the real-time system, evaluating its performance and effectiveness in achieving smooth and accurate trajectory tracking.