Iterative-refinement for action timing discretization

Authors:
Todd W. Neller
Affiliations:
Department of Computer Science, Gettysburg College, Gettysburg, PA
Venue:
Eighteenth national conference on Artificial intelligence
Year:
2002

Citing 6
Cited 0

Depth-first iterative-deepening: an optimal admissible tree search

Artificial Intelligence
Linear-space best-first search

Artificial Intelligence
Artificial intelligence: a modern approach

Artificial intelligence: a modern approach
Robot Motion Planning

Robot Motion Planning
Sensory-motor primitives as a basis for imitation: linking perception to action and biology to robotics

Imitation in animals and artifacts
Simulation-based search for hybrid system control and analysis

Simulation-based search for hybrid system control and analysis

Quantified Score

Hi-index	0.00

Visualization

Abstract

Artificial Intelligence search algorithms search discrete systems. To apply such algorithms to continuous systems, such systems must first be discretized, i.e. approximated as discrete systems. Action-based discretization requires that both action parameters and action timing be discretized. We focus on the problem of action timing discretization.After describing an ε-admissible variant of Korf's recursive best-first search (ε-RBFS), we introduce iterative-refinement ε-admissible recursive best-first search (IR ε-RBFS) which offers significantly better performance for initial time delays between search states over several orders of magnitude. Lack of knowledge of a good time discretization is compensated for by knowledge of a suitable solution cost upper bound.