Nonlinear Adaptive And Optimal Control Strategy For Ship Course

Ship motion has the characteristics of nonlinearity in nature, model uncertainties, anddisturbances are different to measure accurately etc. Thus, the ship motion control belongs tothe theory of nonlinear system control. Besides, the actuator of ship motion control has aseries of hard nonlinear characteristics such as the saturation, dead zone and hysteresis and soon and these characteristics must be considered in the process of controller designed. The shipcourse control is the core of ship automation and it has the direct bearing on themaneuverability, economy, safety of navigation and the fighting capacity. Therefore, it isnecessary to design course control and consider the nonlinear characteristics of the actuatorbased on the uncertainty nonlinear model. It has important theoretical value and practicalsignificance. To this end, based on the methods of nonlinear adaptive control and nonlinearrobust optimal control, the course control problem with characteristics of the actuator isinvestigated. The research work as follows:Firstly, for the realization the state reconstruction and wave filtering of ship coursemotion, a nonlinear observer is designed based on passivity theory. The system model withwave disturbance shaping filter of ship course motion, state observer and an estimation errorsystem of nonlinear equations are given at first. Then the parameter design method of stateobserver is obtained by estimating the passivity of the error system. In order to adapt todifferent sea conditions and navigation conditions, the recursive least squares method isapplied to identify the wave disturbance frequency online. This method makes the passivestate observer parameter adjust automatic. Finally, simulation experiment with the4-DOFmodel of a container ship shows that the proposed controller can realize the wave filtering andstate reconstruction better.Secondly, for the problems of nonlinearity and parameter uncertainty of ship coursemotion, an adaptive backstepping controller is investigated. Furthermore, considering theproposed passive nonlinear state observer, an actual usable output feedback control algorithmis given. To compensate the constant steady-state error and deal with unmodelled dynamics,the integral action is introduced in backstepping and adaptive backstepping. Then theparameter adjustment method is given with reference to PID control. Combine passive stateobserver with backstepping and adaptive backstepping control method proposed in chapter4,the output feedback backstepping and the output feedback adaptive backstepping controller ofship course motion are given, respectively. And we proved that the controlled system is input-state stability. Finally, a dead zone method is employed to avoid the parameter drift inadaptive backstepping control. Simulation results show that the proposed control algorithmimproved the performance and has good adaptability for the uncertainty parameters in the shipcourse motion.Thirdly, for dealing with the problem of steering gear saturation characteristic, ananti-windup backstepping control algorithm is proposed based on the anti-windup andbumpless transfer (AWBT) method. The anti-windup control problem of ship course motionis given at first. Then this problem is extending to investigate anti-windup backsteppingcontrol of a class of uncertainties nonlinear strict-feedback system. The controller designedhas two steps. One step is design adaptive backstepping control without considering the effectof actuator saturation characteristics. Then put the difference of controller output and input asthe input to design anti-windup compensator. This anti-windup backstepping control methodapplied in ship course motion can reduce the effects of steering gear saturation to the systemstability and control performance. Simulation results show that the ship course backsteppingcontroller by employing anti-saturation compensator can reduce the system energyconsumption and improved the control performance effectiveness.Finally, for realizing the nonlinear optimal control of ship course motion, consider theobservation error induced by adding state observer and model parameter uncertainty, abackstepping nonlinear H_∞optimal control algorithm is investigated based on Norrbinmodel The ship course control problem with state observer is analyzed and the local andglobal performance indexes are given. Furthermore, this problem can be reduced to optimalcontrol problem of a class of typical nonlinear strict-feedback system. Backstepping is used totransform the system into a particular form and solve it by using the inverse optimizationmethod. An H_∞interference suppression nonlinear optimal controller is designed. Then anew H_∞optimal controller is investigated that satisfies the requirements of ship coursecontrol. From the simulation results, we can see that nonlinear H_∞optimal controller hasstrong robustness and control performance than LQR method.