Mobile robot control using intelligent agents

The agents' theory offers flexibility for solution of problems whose environment is dynamic and imprecise. The use of the computational intelligence together with the agents' theory seems to be a natural way of providing an agent with intelligence. In this paper we describe the use of intelligent agents, whose intelligence is based on a fuzzy logic system, applied to the control of a robot, simulated by the Khepera simulator [1]. Fuzzy Logic Systems have demonstrated, through the numerous applications in the area, to be an effective procedure for control problems [2][3]. The attitude to be taken at each moment by an agent is defined by a set of fuzzy rules based upon the robot position, its sensor values, distance and angle relative to the target. To prevent the robot from getting stuck by some obstacles, a path memory system was created, forcing the robot to seek new alternatives when it gets trapped [4]. The results obtained demonstrate a successful combination of Computational Intelligence and the Theory of Agents in a control system with ability to avoid deadlock situations.