Research On Self-Organizing Metamorphic Planning For Modular Self-Reconfigurable Multi-Fingered Hands

As an end-effector of a robot manipulator, the hand realizes grasping and dexterous manipulating in complex situation, which has attracted serious attentions in robot community. Based on the synthetic analysis of the state and development of modular self-reconfigurable robots and multi-fingered robot hands, a novel concept named the modular self-reconfigurable robot hand is presented.The intelligent algorithm of configuration topological transformation of modular self-reconfigurable robot hands, self-organizing metamorphosis schemes and realization of visual simulation have been analyzed.Firstly, based on the synthetic analysis of typical module of modular self-reconfigurable robots, cuboid and crystal models of modules are bulit. The models are described by symbolic vector algorithm, and the mathematics model is built. Models of modular self-reconfigurable robot hands are desicrbed by adjoin matrixes and symbolic vector matrixes.Secondly, configuration topologies of modular self-reconfigurable multi- fing -ered hands are described. Configuration topological transformation for modular self-reconfigurable robot hands is programmed. The topological transformation is optimized by Genetic Simulated Annealing and Ant Colony Algorithm. The results show that the connective variations, i.e. connections or disconnections in the self-reconfiguration are reduced, which to improve the efficiency of self-reco -nfiguration for modular self-reconfigurable robot hands.Thirdly, considering basic module motion rules, the distributed control schemes are built by the easy and logical rule, based on the full-discrete and intelligent metamorphic planning method. So, the metamorphic complexity and steps are reduced, and the self-organizing metamorphosis can be obtained.Finally, a visual simulation platform is developed by VC/OpenGL to demonstrate the moving, the climbing over obstructions and the grasping objects.