Development of complex Scenarios and control algorithms for Autonomous Driving Functions (ADFs) in a driving simulator.

The development of Advanced Driver Assistance Systems (ADAS) represents a rapidly evolving field, focused on enhancing vehicle safety and efficiency through the automation of key driving functions, including steering, acceleration, and braking. These systems serve as the foundation for autonomous vehicles, which aim to operate without human intervention through the integration of sensors, control algorithms, and real-time decision making. However, evaluating control strategies in real-world conditions is often costly and impractical, thereby highlighting the need of using simulation environments for rigorous testing and validation. In this context, this thesis presents a simulation framework for evaluating control strategies in autonomous vehicle applications, integrating Simulink for controller design with the CARLA simulator for realistic testing scenarios. Two control methodologies, namely Proportional-Integral-Derivative (PID) control and Nonlinear Model Predictive Control (NMPC), are implemented to control both lateral and longitudinal vehicle dynamics. To improve the interface between the control system and the CARLA simulator, a dispatching function is developed to convert the desired acceleration outputs from the controller into appropriate vehicle input commands, namely throttle and brake commands. This function is designed based on vehicle data collected from the CARLA simulation environment. The proposed framework is tested in highway scenarios featuring both curved and straight road segments, including overtaking maneuvers. Simulation results validate the effectiveness of the control strategies, highlighting their performance and applicability in real-world driving conditions.