Single Neural Network Control Of Ship Heading Considering Dead Zone Input

Ship maneuvering and motion control system is a servo system,which is significantly affected by external interference,and has the characteristics of input saturation,time lag,dead zone and so on.At the same time,the ship motion mathematical model has the characteristics of strong uncertainty and nonlinearity.Therefore,designing a ship maneuvering and motion control algorithm with good control performance has always been a major concern in the field of nautical science and control.In this paper,a ship’s course uncertain nonlinear system with dead zone input characteristics is considered.A single neural network adaptive backstepping control design method is used to design a ship course and course control algorithm.First of all,the dead zone is modeled according to the dead zone characteristics of the steering gear actuator,combined with the Nomoto ship motion mathematical model to establish the ship course control system.Then,design the control algorithm,design the virtual control rate in each step,keep the unknown part of it into the final step design,and use a neural network to approximate all unknown function,and compensate the dead zone to realize the actual control rate and adaptive law of the control algorithm.The stability of the control algorithm is analyzed using the Lyapunov stability theory.Finally,the effectiveness of the design of the control algorithm is verified by Matlab simulation.Compared with the traditional neural network adaptive backstepping algorithm,the Radial Basis Function Neural Network weights and online learning adaptive parameters of the control algorithm in this paper are significantly reduced,which reduces the complexity of control design.The design process of the control algorithm does not require repeated derivation of the virtual control rate,reducing the computational burden of the controller design.Considering the characteristics of the dead zone of the steering gear actuator,the compensation of the dead zone in the controller design process has an important theoretical value and practical meaning.