Dynamic task allocation for robots via auctions

Abstract

We present an auction-based method for the allocation of tasks to a group of robots. The robots operate in a 2D environment for which they have a map. Tasks are locations in the map that have to be visited by the robots. Unexpected obstacles and other delays may prevent a robot from being able to complete its allocated tasks. Therefore tasks not yet achieved are rebid every time a robot accomplishes a task. This provides an opportunity to improve the allocation of the remaining tasks and to reduce the overall task completion time. We present experimental results that we have obtained in simulation using Player/Stage with this task allocation mechanism