A Goal-Oriented Behaviour-Based Control Architecture For Autonomous Mobile Robots Allowing Learning

. When dealing with autonomous robots, the definition of autonomy is important. Autonomous agents require decision-making capabilities in order to perform difficult tasks in a complex environment. In addition, learning from experience is required in order to improve dealing with future tasks. Over the past few years, several different architectures for autonomous mobile robots have been proposed in the literature. Most of them can be classified into two categories: a functionbased one that decomposes the control system into functional modules and a behaviour-based one that decomposes it into task-achieving behaviours. In this paper, an approach for combining these two categories, i.e., combining reactivity as one of the main characteristics of behaviour-based systems with goal-oriented mechanisms as one of the main characteristics of function-based systems, is presented and possibilities of integrating learning strategies are discussed. The paper gives a short survey of existing system..