Decentralized Multi-Agent Pursuit using Deep Reinforcement Learning
arXiv:2010.08193 · doi:10.1109/LRA.2021.3068952
Abstract
Pursuit-evasion is the problem of capturing mobile targets with one or more pursuers. We use deep reinforcement learning for pursuing an omni-directional target with multiple, homogeneous agents that are subject to unicycle kinematic constraints. We use shared experience to train a policy for a given number of pursuers that is executed independently by each agent at run-time. The training benefits from curriculum learning, a sweeping-angle ordering to locally represent neighboring agents and encouraging good formations with reward structure that combines individual and group rewards. Simulated experiments with a reactive evader and up to eight pursuers show that our learning-based approach, with non-holonomic agents, performs on par with classical algorithms with omni-directional agents, and outperforms their non-holonomic adaptations. The learned policy is successfully transferred to the real world in a proof-of-concept demonstration with three motion-constrained pursuer drones.
arXiv commentPublished in RA-L and ICRA
References in corpus (3)
Cited by in corpus (9)
- Learning Multi-Pursuit Evasion for Safe Targeted Navigation of Drones
- Online Planning for Multi-UAV Pursuit-Evasion in Unknown Environments Using Deep Reinforcement Learning
- Toward multi-target self-organizing pursuit in a partially observable Markov game
- ESP: Exploiting Symmetry Prior for Multi-Agent Reinforcement Learning
- A Multi-Agent Approach for Adaptive Finger Cooperation in Learning-based In-Hand Manipulation
- Reinforcement learning reward function in unmanned aerial vehicle control tasks
- ARBoids: Adaptive Residual Reinforcement Learning With Boids Model for Cooperative Multi-USV Target Defense
- Learning cooperative behaviours in adversarial multi-agent systems
- Co-Training an Observer and an Evading Target