# SAGE-LLM: Towards Safe and Generalizable LLM Controller with Fuzzy-CBF Verification and Graph-Structured Knowledge Retrieval for UAV Decision

arXiv: https://arxiv.org/abs/2602.23719 · HTML: https://mlsanity.com/paper/2602.23719/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 32 references that are in the map.
- First posted on arXiv 27 Feb 2026.

## Builds on (most cited first, 21 of 21)

- [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023)
- [Do As I Can, Not As I Say: Grounding Language in Robotic Affordances](https://mlsanity.com/paper/2204.01691/) (2022)
- [PaLM-E: An Embodied Multimodal Language Model](https://mlsanity.com/paper/2303.03378/) (2023)
- [Instruction Tuning for Large Language Models: A Survey](https://mlsanity.com/paper/2308.10792/) (2023)
- [EmbodiedGPT: Vision-Language Pre-Training via Embodied Chain of Thought](https://mlsanity.com/paper/2305.15021/) (2023)
- [DriveMLM: Aligning Multi-Modal Large Language Models with Behavioral Planning States for Autonomous Driving](https://mlsanity.com/paper/2312.09245/) (2023)
- [Agile But Safe: Learning Collision-Free High-Speed Legged Locomotion](https://mlsanity.com/paper/2401.17583/) (2024)
- [A Survey on LoRA of Large Language Models](https://mlsanity.com/paper/2407.11046/) (2024)
- [How to Train Your Neural Control Barrier Function: Learning Safety Filters for Complex Input-Constrained Systems](https://mlsanity.com/paper/2310.15478/) (2023)
- [A Survey of Constraint Formulations in Safe Reinforcement Learning](https://mlsanity.com/paper/2402.02025/) (2024)
- [Embodied AI: From LLMs to World Models](https://mlsanity.com/paper/2509.20021/) (2025)
- [Large Model Empowered Embodied AI: A Survey on Decision-Making and Embodied Learning](https://mlsanity.com/paper/2508.10399/) (2025)
- [Speed Always Wins: A Survey on Efficient Architectures for Large Language Models](https://mlsanity.com/paper/2508.09834/) (2025)
- [In-Context Meta LoRA Generation](https://mlsanity.com/paper/2501.17635/) (2025)
- [Advancing Embodied Agent Security: From Safety Benchmarks to Input Moderation](https://mlsanity.com/paper/2504.15699/) (2025)
- [Enhancing Autonomous Driving Systems with On-Board Deployed Large Language Models](https://mlsanity.com/paper/2504.11514/) (2025)
- [Safe Deep Policy Adaptation](https://mlsanity.com/paper/2310.08602/) (2023)
- [Safe multi-agent motion planning under uncertainty for drones using filtered reinforcement learning](https://mlsanity.com/paper/2311.00063/) (2023)
- [A Survey on Convex Optimization for Guidance and Control of Vehicular Systems](https://mlsanity.com/paper/2311.05115/) (2023)
- [A CBF-Adaptive Control Architecture for Visual Navigation for UAV in the Presence of Uncertainties](https://mlsanity.com/paper/2402.10729/) (2024)
- [Autonomous Decision Making for UAV Cooperative Pursuit-Evasion Game with Reinforcement Learning](https://mlsanity.com/paper/2411.02983/) (2024)

## Abstract (arXiv)

In UAV dynamic decision, complex and variable hazardous factors pose severe challenges to the generalization capability of algorithms. Despite offering semantic understanding and scene generalization, Large Language Models (LLM) lack domain-specific UAV control knowledge and formal safety assurances, restricting their direct applicability. To bridge this gap, this paper proposes a train-free two-layer decision architecture based on LLMs, integrating high-level safety planning with low-level precise control. The framework introduces three key contributions: 1) A fuzzy Control Barrier Function verification mechanism for semantically-augmented actions, providing provable safety certification for LLM outputs. 2) A star-hierarchical graph-based retrieval-augmented generation system, enabling efficient, elastic, and interpretable scene adaptation. 3) Systematic experimental validation in pursuit-evasion scenarios with unknown obstacles and emergent threats, demonstrating that our SAGE-LLM maintains performance while significantly enhancing safety and generalization without online training. The proposed framework demonstrates strong extensibility, suggesting its potential for generalization to broader embodied intelligence systems and safety-critical control domains.