Image-based Visual Servoing Control of Remotely Operated Vehicle for Underwater Pipeline Inspection

This paper presents the development of an image-based visual servoing control system for underwater vehicles, designed to track underwater pipelines using an onboard monocular camera. We propose a novel line-pickup criterion to ensure continuous, smooth tracking when the pipeline changes its direction. An angle projection method is developed to capture the current heading angle relative to the target line and the target angle in the camera view using a 3D camera projection model. A dynamic model of a remotely operated vehicle (ROV) is developed for tracking control design. A nonlinear controller based on nonlinear feedback linearization method is designed for the ROV to track the target angle. A Lyapunov-based stability analysis is provided to prove the local stability of the nonlinear control. The visual servoing control with embedded image process is simulated in an OpenGL environment and implemented on a BlueROV in real-time experiments. The comparisons with different conditions reveal the effect of control parameters on the tracking performance. The experiments have been conducted to test the performance of the model-based nonlinear controller and the PD controller. From the experimental results, the coordination among control signals and state values has validated the effectiveness of this method even when the ROV is affected by the disturbance.