Kinematic Analysis And Parameter Optimization Of The Metamorphic Anthropomorphic Hand

The metamorphic multi-fingered hand developed by Prof Dai differs from other existing robotic hands in the palm structure. The metamorphic palm could change its configuration flexibly that greatly improves the dexterity and adaptability of the multi-fingered hand and provides a new approach for the structure design of humanoid robotic hand. This thesis studies the grasping and operating performances of metamorphic hand and completes the parameter optimization of the newest four-fingered metamorphic hand. The contents and achievements of this thesis are summarized as follows:Position analysis is carried out via loop-closure equations which are constructed based on the geometric constraints of the metamorphic palm. The equations are computed with numerical method and the solution of the metamorphic five-fingered hand is provided to illustrate the effectiveness of the method. The numerical method is accomplished with MATLAB software and could also be used to solve the position analysis of the other metamorphic palms through simply modifying the palm link parameters.A kind of calculation method for the functional workspace of the metamorphic hand is put forward. As a case study the functional workspace of metamorphic five-fingered hand is calculated, which not only validates the method but also shows the augmented effect of metamorphic palm on the functional workspace.The differential kinematic equations of the metamorphic palm and the whole hand are set up by using screw theory, based on which the kinematic manipulability, redundancy and indeterminacy of the robotic hand are derived. Besides, palm manipulability is proposed to assess the effect of palm joint velocities on the velocity of the object.Palmar shape modulation of the metamorphic dexterous hand is assessed with the creative combination of anatomy and biomechanics research on human hand. It is a more intuitive description that establishes the relationship between human palm and the metamorphic robot palm.Parameter optimization model of the robotic hand is set up by choosing the objective function as achieving the most grasp postures which are similar with those of human hand. Then the optimization is completed by means of penalty function method and genetic algorithm. Through the palm mechanism movement, the optimized metamorphic hand could realize multiple objective grasping postures which greatly highlight its palm metamorphic features and humanoid characteristics.