Generating low cost plans under uncertainty and temporal constraints

Abstract

The paper describes a new approach to generating partially ordered plans with durative actions. Both cost and duration of any task is defined through probability distribution. We propose a planner that generates a plan represented in the form of a partially ordered set of tasks. Each task has a set of temporal constraints, a set of probabilities and a set of constant costs. The planning system generates admissible plans satisfying temporal constraints and allows to select the plan of lowest expected cost