Application of The KNN Algorithm in Determining the Orientation of The Probability Area Containing The Ship Position by GPS Systems on Hai Phong Coastal Area

Abstract

The need for navigation for the maritime industry is also very urgent. Today, most commercial boats have receivers and positioning signal receivers, giving an accuracy of several meters. For vessels arriving at, or passing through small canals, the ability to locate with an accuracy of less than 1 meter is necessary. In these cases, the use of DGPS technology is necessary. The determination of ship location and navigation depends on global satellite navigation systems, mainly GPS systems. In maritime practice, the position of the specified vessel is considered the most probable position and will be the center of the circle of probability area containing the ship position. However, the probability area containing the ship position is scalar, the radius of error of the circle of probability area depends on many factors, such as the deviation of geodetic system, the accuracy of the chart. Therefore, the determination of the most probable position with the highest accuracy is a quite complex problem. The demand for data processing is also greater; Machine Learning is thus contributing to solving this problem. In the framework of the article, the authors propose the application of the KNN algorithm to determine the orientation of the probability area containing the ship position with the most probable positions. The objective of this study is to improve efficiency and safety in maneuvering and navigation for sea vessels and testing for Hai Phong coastal area.