A novel scenario of task planning for multi-robot collaboration system

The paper carefully elaborates the current research of the mission planning for multi-robot system in foreign countries. Then introducing a novel method of mission planning which includes task decomposition and task allocation. Tasks divide into decomposable and non-decomposable, only to meet certain conditions, which is decomposable; necessary and sufficient conditions for decomposability of a task with respect to parallel composition and natural projections. This paper mainly researches on the decomposable task and introduces four decomposability conditions. In addition, the task allocation introduced is based on reputation, which is gained from the evaluation of the completion of historical tasks. Task attempts to be assigned to the robot with relatively higher reputation, which can greatly improve the success rate of its mandate implementation. Finally, simulation and experiment´s results have shown that the model can be used in a multi-robot collaboration system, and has good efficiency.