An AUV Docking Approach Based on Image-Based Visual-Servoing and Modified Super-Twisting Sliding Mode Control

In order to realize high accuracy underwater docking, this article proposes an autonomous underwater vehicle (AUV) docking approach based on a monocular camera. LED lights are mounted around the entrance of a docking station to guide the AUV, and then image-based visual servoing (IBVS) scheme is used to provide the desired trajectory to lead the AUV to the docking station without the depth information of feature points. Moreover, consider influences of external disturbances, model uncertainties and the chattering effect, a modified super-twisting sliding mode controller by using the sigmoid function is designed to realize trajectory tracking in docking process. Finally, a fully-actuated AUV named as TY-01 is used in simulations to testify the effectiveness and robustness of the proposed method. Simulation results illustrate that the image error, AUV position error and orientation error both converge to zero to realize underwater docking under model uncertainties and external disturbances, the chattering effect is largely fixed compared with the original super-twisting sliding mode controller.