CLAS: Coordinating Multi-Robot Manipulation with Central Latent Action Spaces

TL;DR

This paper introduces CLAS, a method for coordinating multi-robot manipulation using shared latent action spaces, improving learning efficiency and performance in simulated tasks by enabling better multi-agent coordination.

Contribution

The paper presents a novel approach to multi-robot manipulation that leverages learned shared latent action spaces for improved coordination and efficiency.

Findings

Enhanced sample efficiency over baselines

Improved learning performance in simulated tasks

Effective coordination among multiple agents

Abstract

Multi-robot manipulation tasks involve various control entities that can be separated into dynamically independent parts. A typical example of such real-world tasks is dual-arm manipulation. Learning to naively solve such tasks with reinforcement learning is often unfeasible due to the sample complexity and exploration requirements growing with the dimensionality of the action and state spaces. Instead, we would like to handle such environments as multi-agent systems and have several agents control parts of the whole. However, decentralizing the generation of actions requires coordination across agents through a channel limited to information central to the task. This paper proposes an approach to coordinating multi-robot manipulation through learned latent action spaces that are shared across different agents. We validate our method in simulated multi-robot manipulation tasks and demonstrate improvement over previous baselines in terms of sample efficiency and learning performance.