Survey of Robot Visual Servoing

Abstract

The history and current research of robot visual servoing are reviewed. Robot visual control systems can be categorized according to different criteria. Based on the feedback representation, vision based robot control techniques can be classified into position based control and image based control, which are discussed in detail. The application of neural networks in vision based robot systems is introduced. Feature selection in visual servoing is discussed also. Several advanced problems and the direction in the research of robot vision are pointed out.