BOAs: BackOff adaptive scheme for task allocation with fault tolerance and uncertainty management

Abstract

In this paper, we propose the BackOff Adaptive scheme (BOAs) as a new technique for the automatic allocation of tasks amongst a team of heterogeneous mobile robots. It is an optimal, decentralized decision making scheme that utilizes explicit communication between the agents. A structured and unified framework is also proposed for task specification. This scheme is fault tolerant (to robot malfunctions) and allows for uncertainty in the nature of task specification in terms of the actual number of robots required. Team demography may change without the need for the re-specification of tasks. The adaptive feature in BOAs further improves the flexibility of the team. Realistic simulations are carried out to verify the effectiveness of the scheme. ©2004 IEEE.