Task planning and partial order planning: A domain transformation approach

Abstract

In this paper 1 we introduce techniques of domain transformation for representing and managing task network goals in the framework of partial order planning. A task network planning model, extended to describe external events, is introduced. We prove that it is always possible to express a task network problem in terms of an equivalent problem stated in partial order planning formalism on an appropriate domain. A linear cost technique of domain transformation is described: a given task domain is compiled to generate an equivalent operator based domain which is then submitted to a nonlinear planner. This result shows how to reuse existing partial order planners for solving task network problems. This technique has been successfully demonstrated by the implementation of two TN planners based on domain tranformation for UCPOP and for GRAPHPLAN.

This work has been partially supported under Italian National Research Council C.N.R. “Strategie di controllo innovative per sistemi di Intelligenza Artificiale”