# Cognitive Guardrails for Open-World Decision Making in Autonomous Drone Swarms

arXiv: https://arxiv.org/abs/2505.23576 · HTML: https://mlsanity.com/paper/2505.23576/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Prompt injection and agent security, one of 12 directions in LLM safety and security.
- Builds on 8 of its 34 references that are in the map.
- First posted on arXiv 29 May 2025.

## Placed in

- LLM safety and security / [Prompt injection and agent security](https://mlsanity.com/safety/prompt-injection-and-agent-security/)

## Builds on (most cited first, 8 of 8)

- [VisionLLM: Large Language Model is also an Open-Ended Decoder for Vision-Centric Tasks](https://mlsanity.com/paper/2305.11175/) (2023)
- [The Limits and Potentials of Deep Learning for Robotics](https://mlsanity.com/paper/1804.06557/) (2018)
- [The Next Generation of Human-Drone Partnerships: Co-Designing an Emergency Response System](https://mlsanity.com/paper/2001.03849/) (2020)
- [Extending MAPE-K to support Human-Machine Teaming](https://mlsanity.com/paper/2203.13036/) (2022)
- [Toward Certification of Machine-Learning Systems for Low Criticality Airborne Applications](https://mlsanity.com/paper/2209.13975/) (2022)
- [Ethical Framework for Harnessing the Power of AI in Healthcare and Beyond](https://mlsanity.com/paper/2309.00064/) (2023)
- [Coverage Path Planning for Holonomic UAVs via Uniaxial-Feasible, Gap-Severity Guided Decomposition](https://mlsanity.com/paper/2505.08060/) (2025)
- [Runtime Advocates: A Persona-Driven Framework for Requirements@Runtime Decision Support](https://mlsanity.com/paper/2505.04551/) (2025)

## Built on by (most cited first, 4 of 4)

- [ESARBench: A Benchmark for Agentic UAV Embodied Search and Rescue](https://mlsanity.com/paper/2605.01371/) (2026)
- [An LLM-Agnostic, MAVLink-Based Drone Command and Control Interface and Agentic Harness Using the Model Context Protocol](https://mlsanity.com/paper/2601.15486/) (2026)
- [DisasterBench: A Multimodal Benchmark for UAV-Based Disaster Response in Complex Environments](https://mlsanity.com/paper/2606.06217/) (2026)
- [Agentic AI for Safety-critical Multi-drone Systems: Challenges and Opportunities](https://mlsanity.com/paper/2608.21444/) (2026)

## Abstract (arXiv)

Small Uncrewed Aerial Systems (sUAS) are increasingly deployed as autonomous swarms in search-and-rescue and other disaster-response scenarios. In these settings, they use computer vision (CV) to detect objects of interest and autonomously adapt their missions. However, traditional CV systems often struggle to recognize unfamiliar objects in open-world environments or to infer their relevance for mission planning. To address this, we incorporate large language models (LLMs) to reason about detected objects and their implications. While LLMs can offer valuable insights, they are also prone to hallucinations and may produce incorrect, misleading, or unsafe recommendations. To ensure safe and sensible decision-making under uncertainty, high-level decisions must be governed by cognitive guardrails. This article presents the design, simulation, and real-world integration of these guardrails for sUAS swarms in search-and-rescue missions.