Multi-robot systems with agent-based reinforcement learning: evolution, opportunities and challenges

Multi-agent reinforcement learning for multi-robot systems is a challenging issue in both robotics and artificial intelligence. With the ever increasing interests in theoretical researches and practical applications, currently there have been a lot of efforts towards providing good solutions to this challenge. However, there are still many difficulties in scaling up multi-agent reinforcement learning to multi-robot systems. This paper presents a survey on the evolution, opportunities and challenges of applying agent-based reinforcement learning to multi-robot systems. After reviewing some important advances in this field, some challenging problems and promising research directions are focused on. A concluding remark is made from the perspectives of the authors.