The article presents a comparison of a few controllers of the vessel's course. The mathematical model of the vessel was set as a transfer function with variable coefficients, which depend on vessel speed. We compared a classic PID controller, a PID controller with self-adjusting coefficients, an adaptive controller with implicit reference model, and an adaptive fuzzy controller. In the result of comparison the adaptive fuzzy controller demonstrated the best quality indicators in comparison with other controllers by set of criteria that allows to recommend this controller for implementation in the vessel's course control systems.
The aim of the work is to develop a model of adaptive system of neuro-fuzzy inference based on PI- and PI-FUZZYcontrollers, allowing to simplify, automate and unify the design process of modern automated control systems. To achieve a specific goal, a method for managing a technical object has been developed based on the construction of an adaptive system of neuro-fuzzy inference. As controllers in the system of neuro-fuzzy inference, the classical PI-controller and fuzzy PI-FUZZY-controller were chosen. Interaction between controllers is provided with the help of the hybrid control system developed. The result of interaction of the two models is automatic formation of the basis of fuzzy controller rules based on knowledge of the control object obtained with its control using the classical controller. In the developed adaptive system of neuro-fuzzy inference, error and control signals in the classical model are used as data for building a hybrid network. Error and control signals in the fuzzy model with automatically generated fuzzy inference rules are used as data to verify the hybrid network built in order to detect a fact of its retraining. Thus, during the control of a technical object by means of a hybrid system, the knowledge of an expert in subject domain for adjusting the parameters of the fuzzy controller is completely eliminated, which makes it possible to control difficultly formalizable objects in conditions of uncertainty. To obtain reliable research results, a hybrid control system was developed, consisting of classical and fuzzy models. Numeric values of the error and control signals are obtained at discrete instants of time as a result of interaction of the two models. Special files have been created to build and test a hybrid network in the form of numerical matrices.