Fault Tree Analysis Generation Using Generative Artificial Intelligence with an Autonomy Sensor Usecase

<div>Functional safety forms an important aspect in the design of systems. Its emphasis on the automotive industry has evolved significantly over the years. Till date many methods have been developed to get appropriate fault tree analysis (FTA) for various scenarios and features pertaining to autonomous driving. This article is an attempt to explore the scope of using generative artificial intelligence (GenAI) FTA with the use case of malfunction for the LIDAR sensor in mind. We explore various available open source large language models (LLM) models and then dive deep into one of them to study its responses and provide our analysis. Although the article does not solve the entire problem but has given some guidance or thoughts/results to explore the possibility to train existing LLM through prompt engineering for FTA for any autonomy use case aided with PlantUML tool.</div>