Showing 1–1 of 1 results for author: Kherrour, A

Evaluating Heuristic Search Algorithms in Pathfinding: A Comprehensive Study on Performance Metrics and Domain Parameters

Authors: Aya Kherrour, Marco Robol, Marco Roveri, Paolo Giorgini

Abstract: The paper presents a comprehensive performance evaluation of some heuristic search algorithms in the context of autonomous systems and robotics. The objective of the study is to evaluate and compare the performance of different search algorithms in different problem settings on the pathfinding domain. Experiments give us insight into the behavior of the evaluated heuristic search algorithms, over… ▽ More The paper presents a comprehensive performance evaluation of some heuristic search algorithms in the context of autonomous systems and robotics. The objective of the study is to evaluate and compare the performance of different search algorithms in different problem settings on the pathfinding domain. Experiments give us insight into the behavior of the evaluated heuristic search algorithms, over the variation of different parameters: domain size, obstacle density, and distance between the start and the goal states. Results are then used to design a selection algorithm that, on the basis of problem characteristics, suggests the best search algorithm to use. △ Less

Submitted 3 October, 2023; originally announced October 2023.

Comments: In Proceedings AREA 2023, arXiv:2310.00333

Journal ref: EPTCS 391, 2023, pp. 102-112