Designing Algorithms to Assess Collision Risk in Coastal Waters

Title
Designing Algorithms to Assess Collision Risk in Coastal Waters
Author(s)
전호군; 양찬수
KIOST Author(s)
Yang, Chan Su(양찬수)
Alternative Author(s)
전호군; 양찬수
Publication Year
2019-06-18
Abstract
Current-use CPA (Closest Point of Approach) algorithm has three limitations. Firstly, it does not take into account ship's course-keeping ability. As small ships are more affected by its speed and sea wave, fluctuation of heading is bigger and frequent, and this phenomenon result in fluctuation of Distance of CPA (DCPA) and Time of CPA(TCPA) value. Secondly, ship's maneuverability upon ship size is not considered in the algorithm. As ship is larger, there are more restrictions on direction and speed control, thus avoiding action is larger and slower. Thirdly, since depth are considered, the algorithm cannot identify risk of stranding result from navigation on shallow water. In order to solve these problems, this study aims to design algorithms using ship information including length and draft of ships and water depth as well as course, speed, and position used in the current CPA algorithm. Ship length from Automatic Identification System (AIS) can represent the ship's size, so it is used to subdivide the CPA algorithm according to ship size. Draft information from AIS and depth information from Electronic Nautical Chart (ENC) are used to calculate the ship's stranding risk. First of all, course correction algorithm takes moving average of the cumulative course value for the reliability of the course of small craft. This information helps to identify the risk of collision with small vessels with fast speed and frequent movement change. Secondly, the range of TCPA and DCPA values are graded and risk levels are assigned accordingly. As small vessel has a greater ability to avoid collision risk, TCPA and DCPA values of small vessels are higher than large vessels in same risk level. The stranding risk is determined by whether Under Keel Clearance (UKC) is secured between the ship's draft and the depth of the water in the direction the ship is proceeding. The algorithm processes a vast information of surrounding marine traffic and environment information within a short ti
URI
https://sciwatch.kiost.ac.kr/handle/2020.kiost/22577
Bibliographic Citation
Oceans 2019 Marseille, 2019
Publisher
IEEE Ocean Engineering Society
Type
Conference
Language
English
Files in This Item:
There are no files associated with this item.

qrcode

Items in ScienceWatch@KIOST are protected by copyright, with all rights reserved, unless otherwise indicated.

Browse