# GLAS: Global-to-Local Safe Autonomy Synthesis for Multi-Robot Motion Planning with End-to-End Learning

arXiv: https://arxiv.org/abs/2002.11807 · HTML: https://mlsanity.com/paper/2002.11807/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Feb 2020.

## Built on by (most cited first, 21 of 21)

- [Contraction Theory for Nonlinear Stability Analysis and Learning-based Control: A Tutorial Overview](https://mlsanity.com/paper/2110.00675/) (2021)
- [Distributed Heuristic Multi-Agent Path Finding with Communication](https://mlsanity.com/paper/2106.11365/) (2021)
- [Learning Selective Communication for Multi-Agent Path Finding](https://mlsanity.com/paper/2109.05413/) (2021)
- [Learn to Follow: Decentralized Lifelong Multi-agent Pathfinding via Planning and Learning](https://mlsanity.com/paper/2310.01207/) (2023)
- [Learning Efficient Navigation in Vortical Flow Fields](https://mlsanity.com/paper/2102.10536/) (2021)
- [Learning-based Robust Motion Planning with Guaranteed Stability: A Contraction Theory Approach](https://mlsanity.com/paper/2102.12668/) (2021)
- [Learning Interaction-Aware Trajectory Predictions for Decentralized Multi-Robot Motion Planning in Dynamic Environments](https://mlsanity.com/paper/2102.05382/) (2021)
- [Decentralized Monte Carlo Tree Search for Partially Observable Multi-agent Pathfinding](https://mlsanity.com/paper/2312.15908/) (2023)
- [Learning Control Admissibility Models with Graph Neural Networks for Multi-Agent Navigation](https://mlsanity.com/paper/2210.09378/) (2022)
- [Collision Avoidance and Navigation for a Quadrotor Swarm Using End-to-end Deep Reinforcement Learning](https://mlsanity.com/paper/2309.13285/) (2023)
- [Multi-agent Soft Actor-Critic Based Hybrid Motion Planner for Mobile Robots](https://mlsanity.com/paper/2112.06594/) (2021)
- [Morphological-Symmetry-Equivariant Heterogeneous Graph Neural Network for Robotic Dynamics Learning](https://mlsanity.com/paper/2412.01297/) (2024)
- [Nearest-Neighbor-based Collision Avoidance for Quadrotors via Reinforcement Learning](https://mlsanity.com/paper/2104.14912/) (2021)
- [Local Repair of Neural Networks Using Optimization](https://mlsanity.com/paper/2109.14041/) (2021)
- [Neural-Rendezvous: Provably Robust Guidance and Control to Encounter Interstellar Objects](https://mlsanity.com/paper/2208.04883/) (2022)
- [H-TD2: Hybrid Temporal Difference Learning for Adaptive Urban Taxi Dispatch](https://mlsanity.com/paper/2105.02138/) (2021)
- [Multi-robot Social-aware Cooperative Planning in Pedestrian Environments Using Multi-agent Reinforcement Learning](https://mlsanity.com/paper/2211.15901/) (2022)
- [Multi-Robot Path Planning Combining Heuristics and Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2306.01270/) (2023)
- [CaRT: Certified Safety and Robust Tracking in Learning-based Motion Planning for Multi-Agent Systems](https://mlsanity.com/paper/2307.08602/) (2023)
- [LEARN: Learning End-to-End Aerial Resource-Constrained Multi-Robot Navigation](https://mlsanity.com/paper/2511.17765/) (2025)
- [Trajectory-Level Mode Guidance for Controllable Diffusion-Based Multi-Robot Motion Planning](https://mlsanity.com/paper/2609.36530/) (2026)

## Abstract (arXiv)

We present GLAS: Global-to-Local Autonomy Synthesis, a provably-safe, automated distributed policy generation for multi-robot motion planning. Our approach combines the advantage of centralized planning of avoiding local minima with the advantage of decentralized controllers of scalability and distributed computation. In particular, our synthesized policies only require relative state information of nearby neighbors and obstacles, and compute a provably-safe action. Our approach has three major components: i) we generate demonstration trajectories using a global planner and extract local observations from them, ii) we use deep imitation learning to learn a decentralized policy that can run efficiently online, and iii) we introduce a novel differentiable safety module to ensure collision-free operation, thereby allowing for end-to-end policy training. Our numerical experiments demonstrate that our policies have a 20% higher success rate than optimal reciprocal collision avoidance, ORCA, across a wide range of robot and obstacle densities. We demonstrate our method on an aerial swarm, executing the policy on low-end microcontrollers in real-time.