Image based visual servoing for tele-operated ground vehicles

摘要

Visual servoing has been utilized in advanced mechanics for controlling robots by removing information acquired from the camera, for example, an object distance concerning the object width, length and a reference frame. This paper describes an Image Based Visual Servoing (IBVS), for controlling the manipulator arm of a tele-operated unmanned ground vehicle. In particular, we would discuss a method used for position estimation of objects detected using the visual features extracted from images of a monocular camera fixed on the tele-operated UGV. The camera would function as an eye to hand configuration to the manipulator arm in this proposed method. The configuration has the camera at a fixed point in the world by watching both the target and the robot's end-effector, and is called a send point open loop. Thus the proposed system would need the camera's orientation features and the parameters of the camera to proceed with the calculations. Experimental tests have been carried out to validate the proposed method.