Image-based extension to ant colony algorithm for path finding in gird-based environments

Research output: Contribution to journalArticlepeer-review

Abstract

We propose a technique (IPEAC) that extends the Ant Colony Optimization (ACO) for shortest path finding. In a grid-based environment, when the ACO converges, the optimal path needs to be identified among other emerging paths. Our proposed approach utilizes an image processing algorithm named Connected Component Analysis (CCA). In our implementation, the result of the ACO is an image that models the system elements of source, destination, obstacles, background and agents. This image is fed into CCA which applies a sequence of operators to find the optimal path and calculates its coordinates so that it can be traversed. IPEAC was tested against Dijkstra and A* algorithms. Our experimental work showed that IPEAC is effective and produced an accuracy of 97.8% compared to the A* and 91.8% compared to Dijkstra, however the A* was superior in terms of time efficiency and IPEAC was 60% more efficient that Dijkstra.

Original languageEnglish
Pages (from-to)2853-2867
Number of pages15
JournalInternational Journal of System Assurance Engineering and Management
Volume15
Issue number7
DOIs
StatePublished - 2 Mar 2024

Keywords

  • Ant colony optimization
  • Dijkstra and a* path-finding algorithms
  • Graph algorithms
  • Grid-based path finding
  • Image processing
  • Machine vision
  • Mobile robotics navigation

Fingerprint

Dive into the research topics of 'Image-based extension to ant colony algorithm for path finding in gird-based environments'. Together they form a unique fingerprint.

Cite this