Dexterous manipulation is a challenging and integral task involving a number of sub-problems to be addressed, such as perception, planning and control. Problem representation, which is an essential element in a system that defines what is actually the problem to be considered, determines both the capability of a system and the feasibility of applying such a system in any real tasks. In this talk, I will first introduce our work on designing different representations for the problem of grasp planning, as well as how to use those representations to improve the quality and efficiency of fingertip grasp planning. Based on the proposed grasp optimization framework, I will further discuss the sensing and control during grasp execution and how to enable finger gaiting to maintain grasp stability when external disturbances occur. Additionally, I will introduce our work on learning a grasp manifold to efficiently calculate IK for dexterous hands, as well as how to ensure the motion feasibility of robotic arms during grasp planning. In the end, I will briefly talk about other related projects in my research.
Kaiyu Hang is a postdoctoral associate working with Prof. Aaron M. Dollar at the GRAB lab, Yale University. He received his Ph.D. in Computer Science, specialized in Robotics and Computer Vision, under the supervision of Prof. Danica Kragic in 2016 from KTH Royal Institute of Technology, Stockholm, Sweden. Before joining the GRAB lab, he was a Research Assistant Professor at the Department of Computer Science and Engineering, and a Junior Fellow of the Insititute of Advanced Study, Hong Kong University of Science and Technology.