AI4SSE: ML and LLMs-Enhanced Software and Systems Engineering

Modern software and systems engineering face challenges due to increasing complexity and long project lifecycles. Integrating AI offers transformative opportunities to optimize processes and manage data. Our research group, "AI for Software and Systems Engineering," focuses on using Machine learning and Large Language Models (LLMs) to enhance embedded systems and software engineering.

To ensure the safety and reliability of embedded safety-critical systems, innovative approaches are needed to carry out the validation process in a comprehensive and efficient way. Our research explores the role of AI approach in optimizing traditional testing methods on hardware-in-the-loop (HIL) systems. Specifically, we develop ML and DL models to enhance testing activities during real-time HIL validation in terms of test case generation, execution, and evaluation. By combining innovative AI solutions with HIL testing during the development process, our research aims at improving the safety and robustness of these systems considering the requirements of the ISO 26262 standard.

Our research in software engineering focuses on the application of AI in requirements engineering and architectural design of information systems. The integration of AI language models, i.e., LLMs, into the various phases of the development of software-intensive systems is a growing area of research and offers great potential to support developers in their tasks. The combination of AI techniques and formal methods creates a more systematic and automated approach that significantly improves the quality, efficiency and effectiveness of software development. Research in this area aims to develop innovative, practical and scalable solutions that transform the entire software development process.

In our research group, the following research questions are addressed:

• How can AI-driven methods enhance the efficiency and comprehensiveness of the validation process for embedded safety-related systems in the context of ISO 26262?
• How does the combination of AI language models and formal methods enhance the architectural design of information systems in terms of quality, efficiency, and scalability?

By exploring innovative AI models, and strategies for improving development workflows, our group aims to pave the way for more intelligent, agile software and system development methods.