Visual Docking Against Bubble Noise With 3-D Perception Using Dual-Eye Cameras

Abstract

Recently, many studies have been performed worldwide to extend the persistence of underwater operations by autonomous underwater vehicles. Underwater battery recharging technology is one of the solutions even though challenges still remain. The docking function plays an important role not only in battery recharging but also in other advanced applications, such as intervention. Visual servoing in undersea environments inevitably encounters difficulties in recognizing the environment when captured images are disturbed by noise. This study describes the effective recognition performance and robustness against air bubble disturbances in images captured by a real-time position and orientation (pose) tracking and servoing system using stereo vision for a visual-servoing-type underwater vehicle. The recognition of the vehicle pose based on dynamic images captured by dual video cameras was performed by a real-time multistep genetic algorithm (RM-GA). In previous studies, the docking performance was investigated under the condition that there were no disturbances in the captured images that address image degradation. In this paper, the robustness of the RM-GA against air bubble disturbances was verified through visual servoing and docking experiments in a pool test to confirm that the system can continue to recognize the pose of the 3-D marker and can maintain the desired pose by visual servoing. Then, the effectiveness of the proposed system against real disturbances such as turbidity that may degrade the visibility of the system in the sea was confirmed by conducting the docking experiment in a real sea, having verified the practicality of the proposed method.