Adaptive Grasping And Adjusting Method Of Virtual Hand Gestures With Semantics Information For Virtual Assembly

To improve the grasping efficiency and posture fidelity of virtual hands in virtual assembly processes, a novel adaptive grasping and adjusting method is presented. The attributes and information of assembly parts and assembly gesture of virtual hands are described completely with semantic description models in advance. Basic gestures model base and gestures knowledge base are established to facilitate the generation of hand gestures. According to the spatial relationship between a part and a virtual hand and the semantic description model of the part, the transitions from initial gesture to basic gesture can be calculated. Grasping rules are defined and used to adaptively obtain the final gesture and grasping position of virtual hand. Based on the constraints dynamic-solving algorithm and reasoning of semantic information, the virtual hand is adjusted to the appropriate grasping posture automatically. The validity of this method is verified by the virtual assembly of a reducer.