GUIDE: Guided Updates for In-context Decision Evolution in LLM-Driven Spacecraft Operations
For LLM-based spacecraft operations, GUIDE provides a non-parametric method for cross-episode adaptation, addressing the limitation of static prompting in real-time control.
GUIDE enables LLM-driven spacecraft operations to improve across repeated executions by evolving a structured playbook of natural-language decision rules without weight updates, outperforming static baselines in an adversarial orbital interception task.
Large language models (LLMs) have been proposed as supervisory agents for spacecraft operations, but existing approaches rely on static prompting and do not improve across repeated executions. We introduce \textsc{GUIDE}, a non-parametric policy improvement framework that enables cross-episode adaptation without weight updates by evolving a structured, state-conditioned playbook of natural-language decision rules. A lightweight acting model performs real-time control, while offline reflection updates the playbook from prior trajectories. Evaluated on an adversarial orbital interception task in the Kerbal Space Program Differential Games environment, GUIDE's evolution consistently outperforms static baselines. Results indicate that context evolution in LLM agents functions as policy search over structured decision rules in real-time closed-loop spacecraft interaction.