Contents I. Introduction
With the continuous development of robot technology, the Human-Robot Interaction[1] technology has been paid more and more attention. Nowadays, there are many ways for Human-Robot Interaction. The technology of Human-Robot Interaction based on the data glove is a very active research in recent years. It makes Human-Robot Interaction be accepted and used more easily. It also makes the Human-Robot Interaction be more friendly and human. Fu Yu Jin and Yuan Kui, in the Institute of Automation of Chinese Academy of Sciences, create an improved movement model of handsaccording to the physical constraints of the fingers. It can accurately reproduce the user's hands movements and meet the requirements of the real-time Human-Robot Interaction[2]. Zeng Fen FANG and Liang Bo Lin, in the East China Shipbuilding Institute, analyze the structure and joint of hands and create the virtual hands model based on the link of constraints and the joints of “Father, Son object”. It implements the three dimensional interactive operating environment of gesture synthesis, collision detection and collision treatment based on data glove[3]. Gao Long Qin and Wang Ai Min, in the Mechanical Engineering of Yangzhou University and Instrument Science and Engineering of Southeast University, analyze the physiological structure of hands and create the link between data glove and model of virtual hands according to the motion model of human's hands. Finally, it implements the Human-Robot Interaction based on hands [4]. Not only expensive but the supporting software of data glove are completely closed. The system development of data glove is very complex and the stability and real-time of signals processing is not very good. This is bound to create difficulties for the secondary development of products. It cannot meet the special needs of some users whose hands are disability or finger movements are not flexible.
