Integration and validation of a formula student driverless system

Autonomous driving systems are gaining relevance in the context of motor sport engineering and research. In the context of the Formula Student Driverless (FSD) competition, this thesis addresses the challenge of integrating a functional, rule-oriented driverless system onto the previous season ZUR04 race car. The work builds upon previous isolated developments with the goal of merging and implementing them into a cohesive, operational platform. A modular, hardware-first development strategy was followed, supported by a hybrid methodological approach combining bottom-up engineering, V-Model validation, and agile sprint planning. A test bench closely resembling the car architecture was used to enable early development and verification of subsystems, including safety components, control software, and driverlessspecific interfaces. Key components such as the Emergency Brake System (EBS), Remote Emergency System (RES), and Human-Machine Interface (HMI) were implemented and validated. A steering angle sensor was added to the localization layer to provide essential feedback for vehicle control. The control stack was redesigned using the ROS2 framework to improve modularity and testability. Communication between the Driverless Compute Unit (DCU), Driverless ECU (DV-ECU), and Vehicle Control Unit (VCU) was realised through standardised data interfaces. The result is a functional prototype capable of operating on ZUR04, providing a robust foundation for future integration into the next-generation vehicle ZUR06. While certain features—such as perception, trajectory and mapping remain outside the scope, the thesis successfully establishes all low-level systems necessary for autonomous operation and enables future development through a validated and extensible platform.