Dynamics identification for enhanced haptic display in VR based training platforms

摘要

Current VR systems mainly use geometric models, which has proved to be insufficient to provide the haptic display capability needed in many applications such as surgery training. Physics based dynamic models play a crucial role in this respect, e.g. for realistic haptic display of the operating feel via Virtual Reality (VR) systems. Such physics based models are desirably obtained via experimental identification. However, traditional dynamics identification methods normally require very large sized training data sets, which maybe difficult to meet in practical applications. This paper presents a method for identifying dynamics models using Support Vector Machines (SVM) regression algorithm which is more effective than traditional methods for high dimensional sparse training data. This method can be used as a generic learning machine or as a special learning technique that can make full use of the known dynamics structure knowledge. The experimental results show the application of our method identifying friction models for realistic haptic display.