The forward kinematic modeling of a Stewart platform using NLARX model with wavelet network

Abstract

A Stewart platform manipulator is 6 degrees of freedom parallel robot with a superior performance over serial robots. The forward kinematic problem of a Stewart platform is to determine the actual position and orientation of the movable platform with respect to the base for a given set of leg lengths. The existing analytical methods to solve this problem give multiple solutions, out of which an exact and practical solution has to be selected. A novel method based on non linear autoregressive model with exogenous input (NLARX) using wavelet network is proposed in this paper for the identification of forward kinematic model of a Stewart platform. For a given position of the platform, the leg lengths are estimated using inverse kinematics. This information is used to identify and validate the NLARX model. The simulation results show the ability of the estimated NLARX model to represent the dynamic behavior of the Stewart platform in the forward kinematic mode.