In this research, a aovei self-organizing map (SOM)-based multi-agent architecture is proposed for task assignment coordination and cooperation of a multi-robot system. A set of static or moving targets are randomly distributed in a finite environment (workspace). Every target requires a specific number of robots (that is generally unknown to the robots) to complete a task at its location. Hence, the goai is to dynamically assign a adequate team of robots to every target location or one robot to several target locations with minimal travel distance for each robot. The task is completed when every target has the desired number of robots. The proposed approach integrates the task requirement of robots and the robot motion planning such that the robots can change their direction of movement according to changes in the environment. Unlike most conventional models that are suitable to static environments only, the proposed approach is capable of dealing with non-stationary environments

Keywords