From robotic hands to human hands: a visualization and simulation engine for grasping research

Purpose - Robotic hands are still a long way from matching the grasping and manipulation capability of their human counterparts, but computer simulation may help us understand this disparity. We present our publicly available simulator, and describe our research projects involving the system including the development of a human hand model derived from experimental measurements. Design/methodology/approach - Unlike other simulation systems, our system was built specifically to analyze grasps. It can import a wide variety of robot designs by using standard descriptions of the kinematics and link geometries. Various components support the analysis of grasps, visualization of results, dynamic simulation of grasping tasks, and grasp planning. Findings - The simulator has been used in several grasping research problems and can be used to plan grasps for an actual robot. With the aid of a vision system, we have shown that these grasps can be executed by a robot. Research limitations/implications - We are currently developing methods to handle deformable surfaces, tendon driven models, and non-ideal joints in order to better model human grasping. Practical implications - This work is part of our current project to create a biomechanically realistic human hand model to better understand what features are most important to mimic in the designs of robotic hands. Such a model will also help clinicians better plan reconstructive hand surgeries. Originality/value - We describe our publicly available grasping simulator and review experiments performed with it. The paper demonstrates the usefulness of this system as a tool for grasping research.