Simple optimization techniques for A*-based search

Authors:
Xiaoxun Sun;William Yeoh;Po-An Chen;Sven Koenig
Affiliations:
University of Southern California, Los Angeles, CA;University of Southern California, Los Angeles, CA;University of Southern California, Los Angeles, CA;University of Southern California, Los Angeles, CA
Venue:
Proceedings of The 8th International Conference on Autonomous Agents and Multiagent Systems - Volume 2
Year:
2009

Citing 8
Cited 2

Heuristics: intelligent search strategies for computer problem solving

Heuristics: intelligent search strategies for computer problem solving
Real-time heuristic search

Artificial Intelligence
Speeding up problem solving by abstraction: a graph oriented approach

Artificial Intelligence - Special volume on empirical methods
Performance bounds for planning in unknown terrain

Artificial Intelligence - special issue on planning with uncertainty and incomplete information
Real-time adaptive A*

AAMAS '06 Proceedings of the fifth international joint conference on Autonomous agents and multiagent systems
Speeding up moving-target search

Proceedings of the 6th international joint conference on Autonomous agents and multiagent systems
Generalized Adaptive A*

Proceedings of the 7th international joint conference on Autonomous agents and multiagent systems - Volume 1
Dynamic weighting A* search-based MAP algorithm for Bayesian networks

IJCAI'07 Proceedings of the 20th international joint conference on Artifical intelligence

The increasing cost tree search for optimal multi-agent pathfinding

Artificial Intelligence
Towards rational deployment of multiple heuristics in A*

IJCAI'13 Proceedings of the Twenty-Third international joint conference on Artificial Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we present two simple optimizations that can reduce the number of priority queue operations for A* and its extensions. Basically, when the optimized search algorithms expand a state, they check whether they will expand a successor of the state next. If so, they do not first insert it into the priority queue and then immediately remove it again. These changes might appear to be trivial but are well suited for Generalized Adaptive A*, an extension of A*. Our experimental results indeed show that they speed up Generalized Adaptive A* by up to 30 percent if its priority queue is implemented as a binary heap.