A Conceptual COLREGs-based Obstacle Avoidance Algorithm Implementing Dynamic Path Planning and Collision Risk Assessment

Hasan Uğurlu*, Omar Djecevic, İsmail Çiçek

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

This paper introduces an algorithm for collision avoidance systems intended for Maritime Autonomous Surface Ships. The algorithm takes into account the rules defined by the International Regulations for Preventing Collisions at Sea (COLREGs) and is tailored for real-world maritime environments. Analysis of collision accidents highlights human error and non-compliance with COLREGs as primary contributing factors. Employing the COLREGs as foundational design criteria can mitigate these factors. The algorithm also has the potential to serve as a decision support system on currently manned vessels, enhancing safe navigation. The study proposes a novel rule-based collision avoidance algorithm that adheres to COLREGs, utilizes the Collision Risk Index and ship domain for safety assessment, and combines dynamic path planning for collision-free navigation. Leveraging the Automatic Identification System, which is present on all ships navigating international waters, the algorithm achieves target detection. The algorithm was applied to simulate past ship-to-ship collision incidents, considering the ship kinematics, dynamics, and maneuverability resulting in successful prevention of such accidents through the proposed approach.

Original languageEnglish
Pages (from-to)377-394
Number of pages18
JournalJournal of Eta Maritime Science
Volume12
Issue number4
DOIs
Publication statusPublished - 2024

Bibliographical note

Publisher Copyright:
Copyright© 2024 the Author.

Keywords

  • Collision avoidance
  • Collision Risk Index (CRI)
  • COLREGs
  • Dynamic path planning
  • Ship domain

Fingerprint

Dive into the research topics of 'A Conceptual COLREGs-based Obstacle Avoidance Algorithm Implementing Dynamic Path Planning and Collision Risk Assessment'. Together they form a unique fingerprint.

Cite this