A Visual Servoing Method based on Point Cloud

A visual servoing method based on point cloud is proposed in this paper. This method does not need any image data and simply relies on dense point cloud information, thus is independent of illumination changes. Meanwhile, this method is free of feature matching which is widely employed in traditional visual servoing methods. Experiments on a real robot prove that this proposed method is effective and robust to regular scene as well as the scene in which the object is scarce in texture features.