A SOM-BASED MULTI-AGENT ARCHITECTURE FOR MULTIROBOT SYSTEMS

Abstract

In this paper, a self-organizing map (SOM)-based multi-agent architecture is proposed for multirobot systems. It is capable of controlling a group of mobile robots to complete multiple tasks simultaneously. By cooperative and competitive behaviours, the group of mobile robots can automatically arrange the total task, and dynamically adjust their motion whenever the environment is changed. As an implementation, it can control a group of mobile robots to complete multiple tasks at different locations, such that the desired number of robots will arrive at every target location from arbitrary initial locations. The proposed approach integrates the task requirement of robots and the robot motion planning, such that the robots can start to move before their destinations are finalized. The robot navigation can be dynamically adjusted to guarantee that each target location has the desired number of robots, even under unexpected uncertainties, such as when some robots break down, some robots and/or some tasks are added, or some tasks are changed. Unlike most conventional models that are suitable to static environments only, the proposed approach is capable of dealing with changing environments. In addition, the proposed algorithm can be applied to the path planning of multirobot systems, where a group of robots is coordinated to visit a set of depots. The effectiveness and efficiency of the proposed approach are demonstrated by simulation studies.