Object tracking and visual servoing using features computed from local feature descriptor

In service robot applications, tracking and visual servoing are essential to find objects and position the end-effector of a robot to manipulate an object. In this paper, we propose a high-speed object tracking method based on a window approach and a local feature descriptor, SURF (Speeded-Up Robust Features). The visual servo controller uses geometrical features that are computed directly from the set of SURF interest points, which makes a method robust to the loss of features caused by occlusion or changes in the view point. Furthermore, these features decouple the translations and rotations from the image Jacobian and also keep the object inside the field of view of the camera. Various experiments with a robotic arm equipped with a monocular eye-in-hand camera demonstrate that objects can be grasped safely and stably in the cluttered environment using the proposed method.