A behavior based nonlinear multi-agent framework using a hybrid approach

In this paper, a novel design and implementation of a multi-agent system via a hybrid approach is proposed. First of all, an autonomous agent is insured to be stable by employing the proposed local feedback control besides appropriate switching among well-posed behaviors including obstacle avoidance, target approaching, wandering, manipulator grasping, etc. Next, a hybrid framework of the multi-agent system (HMAS) is established for dealing with multi-agent cooperative problems. The architecture of the inherent hierarchical control is designed consisting of two levels, where the first (lower) level mainly handles the continuous state describing the data acquired from various sensors, whereas the second (higher) level is responsible for providing a well performed supervisory control that can manipulate all available resources to successfully accomplish the underlying mission. In the Advanced Control Laboratory of the Department of Electrical Engineering at National Taiwan University, an actual wheeled mobile robot (WMR) has been developed, named The Treasure Hunter (TTH), which is able to hunt treasures in an unknown environment whereby the effectiveness of the proposed approach can be satisfactorily demonstrated.