Research On Online Speed Optimization For Unmanned Sailboat Via Extremum Seeking Control

In recent years,with the crisis of energy and environment becoming more prominent,unmanned sailboats have gradually become the focus of many maritime industries due to the advantages of reducing navigation costs and eco-friendly sailing.As a kind of clean and intelligent vehicle in the ocean,the advantages of unmanned sailboats are obvious in marine data collection and maritime risk monitoring.Making full use of wind energy to make the sailboat at the optimal speed,so as to achieve maximum safety and efficiency,a reasonable online speed optimization scheme has an important role in improving the overall performance of the sailboat.Therefore,the in-depth research on online speed optimization of unmanned sailboat has a certain theoretical significance and practical value.This paper aims at online speed optimization of unmanned sailboat,taking a 12-m unmanned sailboat as the research object,the traditional speed optimization method is considered to have poor dynamics,modeling errors and steady-state oscillation,combining a feedforward and extremum seeking control(ESC)method,designed the speed optimization scheme,realizes the online speed optimization for unmanned sailboat via ESC.Firstly,to solve the speed optimization problem for unmanned sailboat,combining with the sailing speed polar diagram knowledge and classical perturbation-based extremum seeking control,a classical perturbation-based extremum seeking speed online optimization scheme is proposed,this scheme consists of feedforward and feedback control:the initial sail angle of attack is first computed by a piecewise constant function in the feedforward block,which is equivalent to the initial value given by expert experience,and ensures a small deviation between the sailing speed and the maximum speed;The feedback control uses the ESC to fine-tune the output speed and make it reach the optimal speed,however,there is a problem of steady-state oscillation in this scheme.Secondly,to solve the problem of steady-state oscillation in the classical perturbation-based ESC speed online optimization scheme,a speed optimization scheme of unmanned sailboat via ESC without steady-state oscillation is proposed.In the process of continuously optimizing longitudinal speed,a functional relationship between the excitation signal amplitude and the extremum estimation error is constructed.As the system gradually converges,the amplitude of the excitation signal gradually decreases to zero,thereby eliminating steady-state oscillation and get maximum speed.Thirdly,the basic principles of the traditional sliding mode extremum seeking control were studied.The performance of this algorithm depends on the pre-designed sliding mode surface.When the sliding mode surface occurs,the algorithm performance is not subject to the uncertainty of the parameters and external disturbances,this algorithm will converge to the extreme value of the cost function at a predetermined speed on the sliding mode surface.However,like the classical perturbation-based ESC,the sliding mode ESC speed optimization scheme also has the problem of steady-state oscillation.Aiming at the problem of steady-state oscillation,a switching law is used to combine sliding mode ESC and perturbation-based ESC without steady-state oscillation,and a sliding mode extremum seeking without steady-state oscillation speed optimization scheme is proposed.During the continuous optimization process,a switching law is designed to realize the control transformation,so that the speed optimization system carries out sliding mode ESC in the first convergence phase and perturbation-based ESC without steady-state oscillation in the second steady-state phase,thereby eliminating steady-state oscillation and maintaining a fast convergence rate to reach the optimal speed.Finally,simulation verification was performed based on a 12-m unmanned sailboat model.The results show that the online speed optimization scheme for unmanned sailboat via extremum seeking control can make the speed reach the ideal optimal value,eliminate the steady-state oscillation,and have higher control accuracy and stronger robustness.