A supervisory control method for multi-robot task allocation in urban search and rescue

This paper presents the development of a unique supervisory control architecture for effective task allocation of a heterogeneous multi-robot team in urban search and rescue (USAR) applications. In the proposed approach, the USAR tasks of exploring large unknown cluttered environments and searching for victims are allocated to different robots in the heterogeneous team based on their capabilities. A single human operator is only needed to supervise the team and share tasks with the robots in order to maximize the use of trained operators. Furthermore, the proposed supervisory controller determines the team behavior when faced with robot failures during task execution. Extensive simulated experiments were conducted in USAR-like environments to investigate the performance of the proposed supervisory control method. The results demonstrated that the proposed approach is effective for multi-robot control in USAR applications, and is robust to varying scene scenarios and increasing team size.