Two-stage training algorithm for AI robot soccer

TL;DR

This paper introduces a two-stage training algorithm for heterogeneous multi-agent reinforcement learning in AI robot soccer, enhancing cooperative behavior and performance by balancing individual role and team rewards.

Contribution

The paper proposes a novel two-stage centralized training method that improves learning efficiency and cooperation among heterogeneous agents in multi-agent systems.

Findings

Achieves higher role and team rewards compared to existing methods.

Effectively trains heterogeneous robot soccer teams in simulation.

Enhances cooperative behavior among diverse agents.

Abstract

In multi-agent reinforcement learning, the cooperative learning behavior of agents is very important. In the field of heterogeneous multi-agent reinforcement learning, cooperative behavior among different types of agents in a group is pursued. Learning a joint-action set during centralized training is an attractive way to obtain such cooperative behavior, however, this method brings limited learning performance with heterogeneous agents. To improve the learning performance of heterogeneous agents during centralized training, two-stage heterogeneous centralized training which allows the training of multiple roles of heterogeneous agents is proposed. During training, two training processes are conducted in a series. One of the two stages is to attempt training each agent according to its role, aiming at the maximization of individual role rewards. The other is for training the agents as a whole to make them learn cooperative behaviors while attempting to maximize shared collective rewards, e.g., team rewards. Because these two training processes are conducted in a series in every timestep, agents can learn how to maximize role rewards and team rewards simultaneously. The proposed method is applied to 5 versus 5 AI robot soccer for validation. Simulation results show that the proposed method can train the robots of the robot soccer team effectively, achieving higher role rewards and higher team rewards as compared to other approaches that can be used to solve problems of training cooperative multi-agent.