A Human–Robot Team Knowledge-Enhanced Large Language Model for Fault Analysis in Lunar Surface Exploration

• Published in: Aerospace Vol. 12; no. 4; p. 325
• Main Authors: Wang, Hao, Xue, Shuqi, Zhang, Hongbo, Wang, Chunhui, Fu, Yan
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.04.2025
• Subjects: Astronauts; Case studies; Collaboration; Commercial spacecraft; Datasets; Decision making; Discovery and exploration; Embedding; Failure; Fault location; Graph theory; human–robot collaboration; human–robot dialogue; Knowledge; knowledge graph; Knowledge management; Knowledge representation; Language; large language model; Large language models; Lunar exploration; Lunar surface; Moon; Natural language; Robotics; Robots; Situational awareness; China
• ISSN: 2226-4310; 2226-4310
• DOI: 10.3390/aerospace12040325

Human–robot collaboration for lunar surface exploration requires high safety standards and tedious operational procedures. This process generates extensive task-related data, including various types of faults and influencing factors. However, these data are characteristic of multi-dimensional, time series, and intertwined. Also, prolonged tasks and multi-factor data coupling pose significant challenges for astronauts in achieving safe and efficient fault localization and resolution. In this paper, we propose a method to enhance the base large language models (LLMs) by embedding knowledge graphs (KGs) of lunar surface exploration, thereby assisting astronauts in reasoning about faults during the exploration process. A multi-round dialog dataset is constructed through the knowledge subgraph embedded in the request analysis process. The LLM is fine-tuned using the p-tuning method to develop a specialized LLM suitable for lunar surface exploration. With reference to the situational awareness (SA) theory, multi-level prompts are designed to facilitate multi-round dialogues and aid decision-making. A case study shows that our proposed model exhibits greater expertise and reliability in responding to lunar surface exploration tasks than classical commercial models, such as ChatGPT and GPT-4. The results indicate that our method provides a reliable and efficient aid for astronauts in fault analysis during lunar surface exploration.