Automated Test Case Generation for Satellite FRD Using NLP and Large Language Model

In recent times, the research on the use of Large Language Models (LLMs) for developing software applications has grown exponentially. Generating test cases for satellite Functional Requirement Documents (FRDs) pose a significant challenge due to their complex nature, requiring intricate analysis. Manual methods are time-consuming and error-prone, prompting the need for automated solutions or semi-automated solutions. This work proposes a novel approach to automate test case generation from FRDs using LLMs and Natural Language Processing (NLP). By harnessing the capabilities of LLMs, our system extracts and interprets complex variables and equations, facilitating the automated creation of comprehensive test cases. This approach aims to streamline the satellite testing process, improving efficiency and accuracy while reducing the burden on human analysts. We generate a custom dataset of 10 samples and then benchmark 4 LLMs on the dataset. We open-source the complete codebase for implementation and for further research.