A Novel Quad Harmony Search Algorithm for Grid-Based Path Finding. (9th September 2014)

• Record Type: Journal Article
• Authors: Koceski, Saso; Panov, Stojanche; Koceska, Natasa; Zobel, Pierluigi Beomonte; Durante, Francesco
• Is Part Of: International journal of advanced robotic systems. Volume 11:Number 9(2014)
• Publication Date: 2014-09-09
• Subjects: Heuristic Algorithms -- Artificial Intelligence -- Computational Intelligence -- Optimization -- Path Planning; Robotics -- Periodicals; Robotics; Periodicals; 629.892
• DOI: 10.5772/58875
• Languages: English
• ISSNs: 1729-8806

Abstract

A novel approach to the problem of grid-based path finding has been introduced. The method is a block-based search algorithm, founded on the bases of two algorithms, namely the quad-tree algorithm, which offered a great opportunity for decreasing the time needed to compute the solution, and the harmony search (HS) algorithm, a meta-heuristic algorithm used to obtain the optimal solution. This quad HS algorithm uses the quad-tree decomposition of free space in the grid to mark the free areas and treat them as a single node, which greatly improves the execution. The results of the quad HS algorithm have been compared to other meta-heuristic algorithms, i.e., ant colony, genetic algorithm, particle swarm optimization and simulated annealing, and it was proved to obtain the best results in terms of time and giving the optimal path.