Constraint‐based multi‐agent reinforcement learning for collaborative tasks
Xiumin Shang, Tengyu Xu, Ioannis Karamouzas, Marcelo Kallmann · Computer Animation and Virtual Worlds · 2023
Abstract In order to be successfully executed, collaborative tasks performed by two agents often require a cooperative strategy to be learned. In this work, we propose a constraint‐based multi‐agent reinforcement learning approach called constrained multi‐agent soft actor critic (C‐MSAC) to train control policies for simulated agents performing collaborative multi‐phase tasks. Given a task with phases, the first phases are treated as constraints for the final task phase objective, which is addressed with a centralized training and decentralized execution approach. We highlight our framework on a tray balancing task including two phases: tray lifting and cooperative tray control for target following. We evaluate our proposed approach and compare it against its unconstrained variant (MSAC). The performed comparisons show that C‐MSAC leads to higher success rates, more robust control policies, and better generalization performance.