Kinematic Analysis for Hybrid 2-(6-UPU) Manipulator Using Wavelet Neural Network

This paper addresses forward and inverse kinematics of a specific class of serial-parallel manipulators, known as 2(6-UPU) manipulators. This manipulator composed of two modules which consist of elementary manipulators with the parallel structure of Stewart Platform. At first, the Kinematics Model of the hybrid manipulator is obtained. As there is a highly nonlinear relations between joint variables, and position and orientation of the end effectors, the inverse kinematic problem of these manipulators is quite complicated to solve. In this study, wavelet based neural network (WNN) with its inherent learning ability, is used to solve the inverse kinematic problem. Also, proposed wavelet neural network is applied to approximate the paths of mid and upper plates in circle and spiral trajectories. Finally, the results of simulation show high accurate performance of proposed method.