Movement Flow-Based Visual Servoing for Tracking Trajectories with Occlusions

In this article, we describe a new approach for the tracking of trajectories with occlusions, using image-based control, referred to here as movement flow-based visual servoing. This approach allows the tracking of trajectories specified in the image space and shows a correct behaviour in the 3-D Cartesian space. One of the main contributions of this method is a tracking that is not dependent on time, which affords a better tracking as the system is not subject to time restrictions. To make the tracking of trajectories with occlusions possible, we describe the inclusion of information from the movement flow into an estimator based on the Kalman filter. To verify the correct behaviour of this approach, an eye-in-hand camera system is employed.