Image-Based Visual Servoing for USV under Wave Perturbations

To improve the control accuracy of unmanned surface vehicles (USVs) in specific control tasks, it is necessary to introduce sensors like cameras instead of the Global Positioning System (GPS). This paper considers the problem of image-based visual servoing (IBVS) for USV under the disturbance of waves. Firstly, to deal with image oscillation caused by undesirable roll and pitch motion induced by waves, a perturbation observer is adopted to compensate the effect of attitude changes of USV. Then, to reduce the dimension of the features, a three dimensional moment feature is introduced, and the state space representation of the feature-target interaction is fully defined. The controller is designed using the backstepping idea, and the global asymptotic stability of this method is proved. Eventually, simulation experiments are conducted to demonstrate the effectiveness and superiority of the method compared to traditional IBVS.