A role-based framework for multi-agent teaming

Multi-agent teaming is a key research field of multi-agent systems. BDI (Belief, Desire, and Intension) architecture has been widely used to solve complex problems. The theory of joint behavior has been widely used to solve the team level optimisation problems. Due to the inherent complexity of real-time and dynamic environments, it is often extremely complex and difficult to formally specify the joint behavior of the team a priori. This paper presents a role-based BDI framework to facilitate cooperation and coordination problems. This BDI framework is extended and based on the commercial agent software development environment known as JACK Teams. A real-time 2D simulation environment known as soccerbots has been used to investigate the difficulties of multi-agent teaming. The layered architecture has been used to group the agents' competitive and cooperative behaviors, which can be learned through experience by using the reinforcement learning techniques.