Selected Method Of Artificial Intelligence In Modelling Safe Movement Of Ships
Free (open access)
The main goal in this paper is presented using one of the methods of artificial intelligence, especially neural networks, for determination of the coefficients of state equations of a ship’s safe motion in a horizontal plane. The recurrent optimization network is used to identify parameters of the ship’s dynamics. A structure and the operating principle of the network and results of computer simulation of a ship’s motion along a desired trajectory are presented in this paper as simulation trials using the MATLAB program. Keywords: safe motion, neural network. 1 Introduction The main problem of ship exploitation in sea movement is assuring the right level of navigational safety to a ship. In order to increase the reliability level of sea traffic and the development of its steering, standards have been created [3, 5]. The structure of an automatic safe system of steering a ship must always include an accurate mathematical model of a ship. Many theoretical and practical works have been focused on problems of mathematical modelling and methods of parametric identification were developed especially for safe motion of ships. These methods were very efficient for both linear static and dynamic objects. Recently, an increasing interest has been observed in combining artificial intelligence tools with classical control techniques. Hence neural networks are used to mathematical model a ship’s dynamic. One of the most important tasks during designing of safe automatic control systems is defining an accurate mathematical model of the ship. The results of much research allows one to model both the linear static and dynamic objects. However, a ship as an
safe motion, neural network.