Multi-Agent Robot Systems: Analysis, Classification, Applications, Challenges and Directions

The field of robotics was previously centered on applications involving individual robots. Due to the expansion of high-tech industries and the ease of access to advanced equipment, a new research stream focused on applications involving multiple robots has emerged. With this shift, larger tasks are decomposed and allocated to multiple autonomous robots capable of achieving the expected goals cooperatively. The autonomy to process multiple inputs, make decisions, and perform actions in a collaborative system consisting of multiple robots has led to the introduction of the concept of an "agent." Research in multi-agent robot systems (MARS) involves multiple robot agents. This article reviews publications on MARS, and an analysis of the state-of-the-art concepts and classification of the main lines of research and applications are presented. This study considers area-specific reports over the last ten years and presents some open and unresolved challenges in the field of MARS. Finally, the work concludes with a discussion of the research lines identified in the review and directions concerning the field.