Learning scalable and efficient communication policies for multi-robot collision avoidance

Learning scalable and efficient communication policies for multi-robot collision avoidance

Serra Gomez, A. (TU Delft Learning & Autonomous Control)
Zhu, H. (TU Delft Learning & Autonomous Control)
Ferreira de Brito, B.F. (TU Delft Learning & Autonomous Control)
Böhmer, J.W. (TU Delft Algorithmics)
Alonso-Mora, J. (TU Delft Learning & Autonomous Control)

2023

Decentralized multi-robot systems typically perform coordinated motion planning by constantly broadcasting their intentions to avoid collisions. However, the risk of collision between robots varies as they move and communication may not always be needed. This paper presents an efficient communication method that addresses the problem of “when” and “with whom” to communicate in multi-robot collision avoidance scenarios. In this approach, each robot learns to reason about other robots’ states and considers the risk of future collisions before asking for the trajectory plans of other robots. We introduce a new neural architecture for the learned communication policy which allows our method to be scalable. We evaluate and verify the proposed communication strategy in simulation with up to twelve quadrotors, and present results on the zero-shot generalization/robustness capabilities of the policy in different scenarios. We demonstrate that our policy (learned in a simulated environment) can be successfully transferred to real robots.

Aerial robots
Collision avoidance
Multi-agent reinforcement learning
Multi-robot communication
Multi-robot systems

http://resolver.tudelft.nl/uuid:d0853c0f-0c0f-4ae6-81cf-66ad45f8eb67

https://doi.org/10.1007/s10514-023-10127-3

0929-5593

Autonomous Robots, 47 (8), 1275-1297

Institutional Repository

journal article

© 2023 A. Serra Gomez, H. Zhu, B.F. Ferreira de Brito, J.W. Böhmer, J. Alonso-Mora