This paper aims to enable multi-agent systems to effectively utilize past memories to adapt to novel collaborative tasks in a data-efficient fashion. We propose the Multi-Agent Coordination Skill Database, a repository for storing a collection of coordinated behaviors associated with the key vector distinctive to them. Our Transformer-based skill encoder effectively captures spatio-temporal interactions that contribute to coordination and provide a skill representation unique to each coordinated behavior. By leveraging a small number of demonstrations of the target task, the database allows us to train the policy using a dataset augmented with the retrieved demonstrations. Experimental evaluations clearly demonstrate that our method achieves a significantly higher success rate in push manipulation tasks compared to baseline methods like few-shot imitation learning. Furthermore, we validate the effectiveness of our retrieve-and-learn framework in a real environment using a team of wheeled robots.
翻译:暂无翻译