Abstract
In this paper, we propose a user interface for object manipulation with a multi-fingered robot hand. In order to achieve object manipulation, the user needs to tell the robot about task information such as desired contact points between the manipulated object and fingers. With our system, the user can input this information using an omni-directional camera and laser modules, while the camera which the user holds at some fingertips is regarded as the virtual object for manipulation. The laser beam is reflected on the surface of the finger when the finger is apart from the camera. The system extracts the features of fingers by processing hand images taken by the camera and identifies whether each finger of the user comes in contact with the camera by using a support vector machine. As a result, correct classification rate of 96% has been obtained.
