Tuesday, June 20, 2023, 13:00
Room 01-012, Georges-Köhler-Allee 102, Freiburg 79110, Germany
This thesis proposes a trajectory tracking control system for advanced driver assistance systems (ADAS) using Model Predictive Control (MPC). The proposed controller aims to track a reference trajectory while ensuring safety, comfort, and stability of the vehicle. We use a 2-DOF predictive model to incorporate kinematic constraints of the vehicle, and satisfy path and control constraints to provide a safety margin while vehicle maneuver. The approach relies on the open-source software package acados to achieve satisfactory real-time performance. The thesis also aims to evaluate the performance of the proposed controller using simulation and experimental results. The system is evaluated on several lane change maneuver scenario on highway, and compared with other classical control methods. The effects of non-white feedback noise and system delay are considered in the evaluation, and measures are undertaken to counter them. The simulation results demonstrate the effectiveness and robustness of the proposed system, highlighting its potential for use in future autonomous vehicles.