Decentralized Role Assignment in Multi-Agent Teams via Empirical Game-Theoretic Analysis
This addresses the challenge of coordination in multi-robot systems without explicit communication, though it appears incremental as it applies existing game-theoretic tools to a specific domain.
The paper tackles the problem of decentralized role assignment in multi-agent teams by formulating it as a dynamic game using empirical game theory, enabling robots to choose roles without communication. In simulations of a collaborative planar manipulation scenario, the method allows agents to effectively collaborate and avoid collisions.
We propose a method, based on empirical game theory, for a robot operating as part of a team to choose its role within the team without explicitly communicating with team members, by leveraging its knowledge about the team structure. To do this, we formulate the role assignment problem as a dynamic game, and borrow tools from empirical game-theoretic analysis to analyze such games. Based on this game-theoretic formulation, we propose a distributed controller for each robot to dynamically decide on the best role to take. We demonstrate our method in simulations of a collaborative planar manipulation scenario in which each agent chooses from a set of feedback control policies at each instant. The agents can effectively collaborate without communication to manipulate the object while also avoiding collisions using our method.