Elsevier

Applied Soft Computing

Volume 9, Issue 4, September 2009, Pages 1217-1224
Applied Soft Computing

Prediction of manually controlled vessels’ position and course navigating in narrow waterways using Artificial Neural Networks

https://doi.org/10.1016/j.asoc.2009.03.002Get rights and content

Abstract

Despite modern navigation devices, there are still some problems for navigating of vessels in narrow waterways because of geographical structures and various disturbances. In this study, a guidance and an early warning method by means of predicting three-minute-ahead position of a vessel, especially in the turning points, has been developed for navigating in narrow waterways. The Istanbul Strait has been specifically studied as a model. Since operators in Vessel Traffic Services (VTS) can watch only straight bearing of vessels on VTS panels but especially for turning regions, they have to foresee a risk on time which may result in a disaster. The objective of this study is to predict the future coordinates of a manually controlled vessel using Artificial Neural Networks (ANN).

Artificial Neural Networks have been trained by using position and speed data collected from vessels which navigated manually in the Strait. Three-minute-ahead position of vessels has been predicted by using the trained ANN. Some on-line experiments have been done in Istanbul VTS centre and it has been observed that the method satisfied the goal in especially turning points of the Strait. Hence the proposed method could be utilized for warning system by VTS operators and guidance system by vessel crew.

Introduction

Turkey is an active actor in the international sea transportation sector and the Straits (Bosphorus and Dardanelles) are at the position of being a natural waterway between the Black Sea and the Mediterranean. Vessels carrying dangerous cargo in high volumes pass through the Turkish Straits where highly populated metropolitan areas are located in very close distance. The vessel traffic in Turkish Straits is 4 times higher than the Panama Canal and 3 times higher than the Suez Canal [1].

Over 400 accidents have occurred in the Istanbul Strait in last 60 years. The statistics show that 80% of accidents are the result of human operator faults. “Selvi” and “Yenikoy” are some turning regions where most violations occur in Istanbul Strait and courses of some vessels, passing from South to North, are shown in Fig. 1, Fig. 2.

After Nassia accident in 1994 on the Strait, it has been largely accepted that, if serious security precautions are not taken, besides financial losses, with another big vessel explosion Istanbul's population of 12 million and history of 3000 years, will face lethal danger.

Even though all vessels have modern navigation devices, use of autopilot is not allowed in the Straits. Vessels have to be manually controlled according to rules. A human operator (helmsman) controls a vessel as a continuous system manually according to captain's command against random disturbances and their behaviour affects the course and coordinates of the vessel [14]. On the other hand, despite modern navigation devices, there are still some problems for navigating of vessels in narrow waterways because of geographical structures and various disturbances [9].

In this study, a guidance and a warning method has been developed for navigation in narrow waterways. The Istanbul Strait has been specifically studied as a model. The basis of this study is to predict the future coordinates of a manually controlled vessel using Artificial Neural Networks (ANN). The method could be utilized for warning system by Vessel Traffic Services (VTS) operators and guidance system by vessel crew.

There are studies appeared in the literature attempting to obtain intelligent models such as ANN and Fuzzy Logic models by learning the human operator actions for prediction purposes [15], [21]. It should also be mentioned here that Ebada [3] used Neural Networks to predict ship turning maneuver and Nicolau et al. [13] predicted the wave influence on the yaw motion of a ship. A comprehensive literature survey does not reveal any system like the suggested one in this paper that could be employed on the VTS systems. Artificial Neural Networks have been trained by using position and speed data collected from vessels which navigated manually in the Strait and they included effects of environmental conditions and geographical characteristics of the Strait.

In practice, since the vessel has high inertia and slow maneuver response, sampling period is selected as 1 min. Although the permitted speed is theoretically 10 knots for navigation in the Strait, if the traffic is convenient, especially in one way navigation, permitted speed can be up to maximum of 15 knots. So, the vessel will move forward approximately 420 m in 1 min on a straight course. Displacement will be 1260 m in 3 min. VTS operators can watch only straight bearing of vessels on VTS panels but especially for turning regions, they have to foresee a risk on time which may result with a disaster because of negligence of the vessel crew. Therefore, it has been decided that the prediction of three-minute-ahead position of vessels by using ANN will be very important. Some experiments have been done in Istanbul VTS centre and it has been observed that the method satisfied the goal in especially turning points of the Strait.

Section snippets

Istanbul Strait and navigation

City of Istanbul, divided into two sides by the Strait, has a population of more than 12 million people and hosts a historical and cultural heritage of 3000 years. Most of the population and historical artifacts are mainly located by the Strait fronts. Furthermore, approximately one and a half million people benefit from sea transportation and thus are on the sea every day. The number of daily passages of sea vehicles between two continents reaches up to 2000. The number of passages in the

Vessel Traffic Services

According to the international maritime literature Vessel Traffic Services (VTS) is a project prepared to maintain a safe vessel passage in the Turkish Straits and started to be implemented as of January 1, 2004. There are eight stations in the Istanbul Strait and five stations in the Dardanelles in order to derive benefit from the VTS system and many vessels can be monitored from these stations with VTS. Works of securing the navigation of vessels passing through within their own Traffic

Artificial Neural Networks

Modeling of non-linear systems is far more difficult than linear systems. The disturbances influencing the system make modeling task even more difficult. Scientists have been studying non-linear system modeling for years and they have succeeded in teaching non-linear system dynamics to Artificial Neural Networks without any mathematical modeling [10], [11], [17].

Neural Networks, with their remarkable ability to learn complicated relations from imprecise data, can be used to extract patterns and

Training of ANN

Latitude and longitude (x, y), course (ψ), speed (u) and time (t) data have been collected from GPS of four vessels with 1 min sampling period to train Artificial Neural Networks. Then the trained ANN has been used to predict x(t + 1), y(t + 1), x(t + 2), y(t + 2) and x(t + 3), y(t + 3) positions. Data from vessels, navigated to south and north, have been used to determine model structures. Although the course is stable on the reference line between two turning points, vessels can change its course if it is

Prediction with trained ANN (TANN)

After training was completed, input–output vectors have been modified as explained above for prediction step with ANN model, as shown in Fig. 8.

For on-line prediction purpose, courses of different vessels on the Istanbul Strait have been observed by using VTS system. Position data of a tanker, namely 244 m (NS Challenger) were collected from VTS system in 1 min sampling period. Artificial Neural Network was retrained after receiving new data and predicted one-, two- and three-minute-ahead of

Conclusion and remarks

In this study, a guidance and an early warning method by means of predicting three-minute-ahead position of a vessel has been developed for navigating in narrow waterways. The Istanbul Strait has been specifically studied as a model. The basis of this study is to predict the future coordinates of a manually controlled vessel using Artificial Neural Networks. A Straight bearing line is shown on the VTS operator panels and it represents vessels’ course but it is not useful especially on the

References (21)

  • P. Singh et al.

    Suitability of different neural networks in daily flow forecasting

    Applied Soft Computing

    (2007)
  • G. Cerit, Deniz Guvenlik Yonetimi ve Turk Bogazlari, Dokuz Eylul Universitesi Deniz Isletmeciligi ve Yonetimi...
  • H.B. Demuth, M. Beale, Neural Network Toolbox for use with MATLAB, User's Guide Version 3, The Math Works Inc.,...
  • A. Ebada, M.A. Maksoud, Prediction of Ship Turning Manoeuvre Using Artificial Neural Networks, University Duisburg,...
  • Y.M. Enab

    Intelligent controller design for the ship steering problem

    IEE

    (1996)
  • S. Ertugrul et al.

    Neuor-fuzzy controller design via modeling human operator actions

    Journal of Intelligent Fuzzy Systems

    (2005)
  • M.T. Hagan et al.

    Neural Network Design

    (1996)
  • M.T. Hagan et al.

    An introduction to the use of neural networks in control systems

    International Journal of Robust Nonlinear Control

    (2002)
  • IALA, International Association of Marine aids to Navigation and Lighthouse Authorities, Guideline No: 45, On staffing...
  • Y. Kondratenko, S. Sydorenko, Automation of Decision Making in Uncertainty: Navigation in Narrowness and Channels,...
There are more references available in the full text version of this article.

Cited by (0)

View full text