This paper presents a novel computation engine designed to enhance the capabilities of Large Language Models (LLMs) to support the initial phases of spacecraft design. With the integration of a Python-based computation engine, we address some of the inherent limitations of LLMs, such as the lack of computational ability and difficulty in tracing decisionmaking processes back to specific data inputs. The computation engine is supported by OpenAI’s API models and the tool call feature to enable accurate, and traceable decision-making within spacecraft system configurations, focusing on missions involving Earth observation using CubeSats in Sun-Synchronous orbits. Through experimental validation across various simulated mission scenarios, the engine demonstrates significant improvements in reducing output uniformity and enhancing stability in design parameter predictions. The system’s design allows for future expansions and adaptability to include a broader range of spacecraft types and mission objectives, marking a significant step forward in the application of AI in aerospace engineering.
«
This paper presents a novel computation engine designed to enhance the capabilities of Large Language Models (LLMs) to support the initial phases of spacecraft design. With the integration of a Python-based computation engine, we address some of the inherent limitations of LLMs, such as the lack of computational ability and difficulty in tracing decisionmaking processes back to specific data inputs. The computation engine is supported by OpenAI’s API models and the tool call feature to enable ac...
»