Expressive Planning and Explicit Knowledge

Planning with Incomplete Information for Robot Problems

Volume

Planning with Incomplete Information for Robot Problems

Contents

In this paper, we explore the implications and interactions of three common expressive extensions to classical planning: conditional plans, context-dependent actions, and nondeterministic action outcomes. All of these extensions have appeared in recent work, sometimes in conjunction, but the semantics of the combination has not really been addressed, save in the probabilistic planner C-BURIDAN (Draper, Hanks, and Weld 1993). We present a new planning language, WCPL, providing a unified treatment and a coherent semantics for conditional planning with context-dependent and nondeterministic actions, including a principled treatment of the distinction between ground truth and the planner’s information state.