University of Strathclyde website
Digital Collections - University of Strathclyde Library
Search Results Previous Searches E-Shelf
Login End Session
Guest
Search 'System Number= 000004495' in 'General Silo' Collection [ Sorted by: Name/Title ] Refine search
Table view Full view
Record 1 of 1 1
Add to E-Shelf
e-item icon
Link(s)
PDF of thesis T14285 PDF of thesis T14285 - (1 M)
Title Heuristically guided constraint satisfaction for AI planning / by Mark Judge.
Name Judge, Mark. .
Abstract Standard Planning Domain Definition Language (PDDL) planning problemdefinitions may be reformulated and solved using alternative techniques.One such paradigm, Constraint Programming (CP), has successfullybeen used in various planner architectures in recent years.The efficacy of a given constraint reformulation depends on the encodingmethod, search technique(s) employed, and the consequent amount ofpropagation.Despite advances in this area, constraint based planners have failed tomatch the performance of other approaches. One reason for this is thatthe structural information that is implicit in the domain / problem instancedescription is lost in the reformulation process.Hence, to achieve better performance, a constraint based planner shouldhave an appropriate encoding and a search procedure informed by thestructure of the original problem. This thesis describes a system that aimsto improve planner performance by employing such methods.The planner uses a table-constraint encoding together with a new variable/ value ordering heuristic. This ordered, goal oriented guidance reducesthe search space and better directs the search focus.The system also makes use of novel meta variable techniques for goallocking and resource assignment. These improve propagation and prunethe search space.This CP based planning architecture is shown to perform well on a range ofproblem instances taken from recent International Planning Competitions(IPCs).
Publication date 2015.
Name University of Strathclyde. Dept. of Computer and Information Sciences.
Thesis note Thesis Ph. D University of Strathclyde 2015 T14285

Powered by Digitool Contact us Electronic Library Services Library Home