As yet, few general control methods to generate cooperation among multiple robots have been reported. So this paper presents a cooperative control method for multiple mobile robots with local sensors. The objective of the presented method is to present how to organize the cooperative system for multiple mobile robots, generally how to design control logics for a single robot and how to recover errors.
The method consists of two sub-methods. The one is a method to design global control plans for many robots, i.e., for task allocation plans, task change plans and so on in order to effectively execute a goal-task. The other is a method to control a single robot in the cooperative system and involves ways to design control logics and to recover errors. To effectively design control logics, ways to integrate sensor information and to generate a control command are presented. The former is to merge similar sensor states to one by using 'AND/OR' logics and the latter is to eliminate logical conflicts by means of transition firing sequence in Petri nets. And in order to recover errors that generally may occur, ways to design reference model and to change control commands based on the control model are introduced.
To demonstrate the effectiveness of these sub-methods, the object-searching task is introduced. This task is to search an object like a box by two robots and consists of two sub-tasks, i.e., a wall tracking task and a robot tracking task. Simulation results for the object-searching task, the wall tracking task and the error recovery methods are presented to show the effectiveness of the method.
