Grasp Identification and Multi-Finger Haptic Feedback for Virtual Assembly

The sense of touch is an important way for humans to feel the world. It is very important to provide realistic haptic feedback in virtual assembly applications as to enhancing immersion experience and improving efficiency. This paper presents a novel approach for grasp identification and multi-finger haptic feedback for virtual assembly. Firstly, the Voxmap-PointShell (VPS) algorithm is adapted and utilized to detect collisions between a dexterous virtual hand and a mechanical component or between two mechanical components, and collision detection results are used to guide the motion of a virtual hand. Then collision forces at collision points are calculated (using Hooke Law), classified and converted. Finally, forces received at fingertips of a virtual hand are used to identify whether or not a virtual hand can grasp a mechanical component, and are mapped to exert forces at user’s fingertips with a CyberGrasp force feedback system. Our approach has been incorporated and verified in a CAVE virtual environment.