ABSTRACT
This chapter introduces mathematical formulation of the design problem of robot behaviour systems in human-robot systems. The human-robot interactions are considered in a multi-agent framework, where all intelligent entities are regarded as agents. The first half of this chapter discusses the general multi-agent model, the encoding of different modes of human-robot interactions into the model, and the features of human-robot systems that need to be considered during the design. The second half of this chapter overviews the methodologies for robot behaviour design, which can follow either a model-based approach or a model-free approach. In particular, a model-based approach which consists of a “see-think-do” architecture is discussed. A model-free approach which takes the advantage of statistical learning is also discussed. The design methodologies will be further elaborated in Part II (Theory) of the book. The application of those methods will be discussed in Part III (Applications) of the book.
