Arrow Research search
Back to ICAPS

ICAPS 2018

Plan Relaxation via Action Debinding and Deordering

Conference Paper Main Technical Track Artificial Intelligence · Automated Planning and Scheduling

Abstract

While seminal work has studied the problem of relaxing the ordering of a plan’s actions, less attention has been given to the problem of relaxing and modifying a plan’s variable bindings. This paper studies the problem of relaxing a plan into a partial plan which specifies which operators must be executed, but need not completely specify their order or variable bindings. While partial plans can provide an agent with additional flexibility and robustness at execution time, many operations over partial plans are intractable. This paper tackles this problem by proposing and empirically evaluating a fixed-parameter tractable algorithm which searches for tractable, flexible partial plans.

Authors

Keywords

  • planning and scheduling
  • partial-order planning

Context

Venue
International Conference on Automated Planning and Scheduling
Archive span
1990-2024
Indexed papers
1573
Paper id
254983529250292886
v2026.09.13