Representing and Solving Temporal Planning Problems

  • R. James Firby
  • Drew McDermott
Chapter
Part of the Symbolic Computation book series (SYMBOLIC)

Abstract

AI planning research has generally been concerned with the advantages and disadvantages of representing temporal planning information as a partially ordered task network. Such a network represents a growing plan as a set of tasks that become more completely temporally ordered as the plan is elaborated. A partially ordered task network is an attractive plan representation because it maintains the flexibility to avoid or take advantage of unexpected task interactions by simply adding appropriate ordering constraints between existing tasks. Unfortunately, a partially ordered network has difficulty representing tasks with properties that depend on precisely what other tasks come before them. How can one keep track of a bank balance in the face of unordered withdrawals and deposits? This paper discusses work done at Yale by Dean and Miller that addresses both the problem of building and maintaining a partially ordered task network and the problem of reasoning about tasks that have order-dependent properties.

The Time Map Manager designed by Dean is a comprehensive system for building a predicate calculus database of assertions that may be true over only certain intervals of time. The database is augmented with a mechanism for representing assertions that persist until contradicted (or “clipped”) and with inference rules that assert new facts whenever particular conjunctions of facts are true at the same time. This database can be used to represent tasks and their effects and to dynamically order unordered tasks as the need arises. The Heuristic Task Scheduler of Miller is an algorithm for finding a consistent temporal order for a set of tasks that have context dependent properties. The tasks to be ordered may have varying durations, varying effects and may contain loops. In addition, tasks may use and alter common resources (such as bank balances) in a wide variety of ways. Constraints on possible orders for the tasks take the form of restrictions on temporal order, task deadlines, and competition for resoursces.

Keywords

Milling Tate Prefix Sonar 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag New York Inc. 1987

Authors and Affiliations

  • R. James Firby
    • 1
  • Drew McDermott
    • 1
  1. 1.Department of Computer ScienceYale UniversityNew HavenUSA

Personalised recommendations