Robot hands play an important role in the interaction between robots and the environment, and the precision and complexity of their tasks in work production are becoming higher and higher. However, because the traditional manipulator has too many driving components, complex control, and a lack of versatility, it is difficult to solve the contradiction between the degrees of freedom, weight, flexibility, and grasping ability. The existing manipulator has difficulty meeting the diversified requirements of a simple structure, a large grasping force, and the ability to automatically adapt to shape when grasping an object. To solve this problem, we designed a kind of underactuated manipulator with a simple structure and strong generality based on the metamorphic mechanism principle. First, the mechanism of the manipulator was designed on the basis of the metamorphic mechanism principle, and a kinematics analysis was carried out. Then, the genetic algorithm was used to optimize the size parameters of the manipulator finger structure. Finally, for different shapes of objects, the design of the control circuit binding force feedback control was carried out with a grasping experiment. The experimental results show that the manipulator has simple control and can grasp objects of different sizes, positions, and shapes.
Keywords: kinematic analysis; manipulator; metamorphic mechanism; underactuated finger mechanism.