Open AccessMethods of ship trajectory data processing for applying artificial neural network in port area
Author(s) -
Kwang Ii Kim,
Keon Myung Lee,
Jang Young Ahn
Publication year - 2018
Publication title -
international journal of engineering and technology
Language(s) - English
Resource type - Journals
ISSN - 2227-524X
DOI - 10.14419/ijet.v7i2.12.11112
Subject(s) - categorical variable , artificial neural network , trajectory , computer science , traffic flow (computer networking) , variable (mathematics) , line (geometry) , variance (accounting) , real time computing , floating car data , artificial intelligence , data mining , simulation , engineering , machine learning , transport engineering , traffic congestion , mathematics , computer network , mathematical analysis , physics , geometry , accounting , astronomy , business
Background/Objectives: In Vessel Traffic Service (VTS), prediction of the flow of vessel traffic is essential to serve safety information and control ship traffic. However, it is difficult to predict a ship’s speed due to many external forces and environmental conditions. This study proposes a data processing method to convert ship speed data to categorical data by dividing ship navigating routes into several gate lines.Methods/Statistical analysis: A ship’s trajectory is converted to each route’s gate line speed. To determine the gate line speed, we convertedthe previous and subsequent gate line speeds into category data. The input and output category data were applied to a multilayer perceptron network using as input variablesthe previous speed variance category, ship type, and ship length, and as output variable the subsequent speed variance.Findings: These results are useful because categorical data can be applied to various neural network models. As a result of the conducted experiments, the accuracy of the model improved when many gate lines are included.Improvements/Applications: The study results can be applied topredict ship traffic flow for VTS operators.