Abstract

Reasoning About Actions and Plans discusses approaches to a number of the more challenging problems in reasoning about the future and forming plans of action to achieve their goals. Reasoning about actions and plans can be seen as fundamental to the development of intelligent machines that are capable of dealing effectively with real-world problems.

This book comprises 17 chapters, with the first delving into the semantics of STRIPS. The following chapters then discuss a theory of plans; formulating multiagent, dynamic-world problems in the classical planning framework; and a representation of parallel activity based on events, structure, and causality. Other chapters cover branching regular expressions and multi-agent plans; a representation of action and belief for automatic planning systems; possible worlds planning; and intractability and time-dependent planning. The remaining chapters discuss goal structure, holding periods and "clouds"; a model of plan inference that distinguishes between the beliefs of actors and observers; persistence, intention, and commitment; the context-sensitivity of belief and desire; the doxastic theory of intention; an architecture for intelligent reactive systems; and abstract reasoning as emergent from concrete activity. This book will be of interest to practitioners in the fields of cognition and artificial intelligence.