English

LEARN: Learning End-to-End Aerial Resource-Constrained Multi-Robot Navigation

Robotics 2025-11-25 v1 Machine Learning Multiagent Systems

Abstract

Nano-UAV teams offer great agility yet face severe navigation challenges due to constrained onboard sensing, communication, and computation. Existing approaches rely on high-resolution vision or compute-intensive planners, rendering them infeasible for these platforms. We introduce LEARN, a lightweight, two-stage safety-guided reinforcement learning (RL) framework for multi-UAV navigation in cluttered spaces. Our system combines low-resolution Time-of-Flight (ToF) sensors and a simple motion planner with a compact, attention-based RL policy. In simulation, LEARN outperforms two state-of-the-art planners by 10%10\% while using substantially fewer resources. We demonstrate LEARN's viability on six Crazyflie quadrotors, achieving fully onboard flight in diverse indoor and outdoor environments at speeds up to 2.0m/s2.0 m/s and traversing 0.2m0.2 m gaps.

Keywords

Cite

@article{arxiv.2511.17765,
  title  = {LEARN: Learning End-to-End Aerial Resource-Constrained Multi-Robot Navigation},
  author = {Darren Chiu and Zhehui Huang and Ruohai Ge and Gaurav S. Sukhatme},
  journal= {arXiv preprint arXiv:2511.17765},
  year   = {2025}
}

Comments

20 pages, 15 figures

R2 v1 2026-07-01T07:49:44.242Z