I am working with Professor Nancy Pollard at the Computer Graphics Lab at Carnegie Mellon University as a postdoctoral fellow. My current researches focuses on humanoid behavior of robots, especially, on human grasping and manipulating objects. A variety of humanlike robot hands have been constructed, but it remains difficult to control these hands in a dexterous way. One challenge is grasp synthesis, where we wish to place the hand and control its shape to successfully grasp a given object [1]. We derived a data-driven approach to grasp synthesis that treats grasping as a shape matching problem. We begin with a database of grasp examples. Given a model of a new object to be grasped (the query), shape features of the object are compared to shape features of hand poses in these examples in order to identify candidate grasps. For effective retrieval, we develop a novel shape matching algorithm that can accommodate the sparse shape information associated with hand pose and that considers relative placements of contact points and normals, which are important for grasp function. We illustrate our approach with examples using a model of the human hand. Our results show that this algorithm can find very good matches between the hand and the object surface. The developed approach can be applied to robot hands and also animated characters in games and virtual environments. [Presentation Slides] [Poster] [Movie]

I received my Ph.D. in System Information Engineering, from Kagoshima University, Japan. Dissertation: A Study on Object Grasp with Multi-fingered Robot Hand [3][4][5]. In the dissertation, I proposed an analytical approach for planning finger position of a 2D object with a multi-fingered hand. At first, a method to obtain a combination of object edges to be used for grasping is derived. Then the Graspable Finger Position Region (GFPR) on a combination of edges is defined where the object can be held successfully. It is shown that the region is bounded by plural boundary hyperplanes in the finger position space. Two propositions for analytically and exactly obtaining the GFPR are proposed. Furthermore, an algorithm to find the stable GFPR that contains the largest inscribed hypersphere of GFPR and has the larger volume is proposed. [Presentation Slides]

I have worked on a training system project for hemiplegics lower extremity [6] as a research student in Kagoshima University of Japan. We collaborated with orthopedic doctors in the hospital. For this project, I established kinematics model of human lower extremity and derived a novel analytical method for inverse kinematics calculation for human leg motion.

I have eight years of experience in vehicle research and design. Before I went to Japan, I worked in the Department of Research and Development, Tianjin Construction Machinery Company, China. I had been involved with the development and evaluation of a variety of construction equipment and accumulated a lot of knowledge and experience in machine design [7][8].

Resume

Contact

Recent Pictures

Links