Library definition for an automotive ECU API layer using Model Based approach - Memory Management & OBD

This thesis, conducted in collaboration with Metatron S.p.A., presents a model-based approach for expanding and redefining an API layer for automotive Electronic Control Units (ECUs), focussing on two key aspects: memory management and on-board diagnostics (OBD). The first part of this work addresses memory management, where a robust, multi-module strategy was developed and implemented to efficiently manage the non-volatile memory (NVRAM). The found solution ensures the secure storage and retrieval of safety-critical data, essential for the reliable operation of automotive systems. The second section introduces a novel OBD strategy, designed to detect, validate, store, and communicate faults in compliance with current emission regulations and heavy-duty vehicle protocols. The designed strategy integrates with the memory management system, leveraging stored error data to support diagnostic accuracy and reliability. By combining these two approaches, the work contributes to the development of a comprehensive ECU API layer that enhances both system safety and diagnostic capabilities. Throughout the project, we updated pre-existing code, elaborated new code from scratch, reduced complexity, and improved usability, with a strong focus on modularity and efficiency. These improvements will create a solid foundation for future developments, enabling easier integration of new features and ensuring scalability for evolving ECU requirements.