Neural network solution for the forward kinematics problem of a redundant hydraulic shoulder

In this paper, a neural network based solution for forward kinematics analysis of a hydraulic shoulder is presented. The shoulder has three degrees of freedom rotational motion produced by four hydraulic cylinders which makes it a redundant parallel robotic shoulder. Unlike the serial robots, forward kinematics problem of parallel robots is not easily solved because of the nonlinearity and complexity of the parallel robot's kinematic equations and there are several solutions which are almost impossible to be found analytically. The neural network used in this paper is of the feedforward net type and a multi-layer back propagation procedure is utilized to train the network. A simulation study is performed using two types of trajectories to illustrate the advantages of the proposed method in solving the forward kinematics problem of the redundant mechanism. The results show the method provides a fast solution and good tracking performance.

I. INTRODUCTION

After a motion simulator with parallel kinematic chains was introduced in 1965 (D. Stewart) [1], parallel manipulators received more and more attention because of their high stiffness, high speed, high accuracy, compact and high carrying capability [2]. They have been used widely in the fields of motion simulators, force/torque sensors, compliance devices, medical devices and machine tools [3], [4].