Identification of Mimo dynamic system using inverse Mimo Neural Narx model

Abstract

This paper investigates the application of proposed neural MIMO NARX model toa nonlinear 2-axes pneumatic artificial muscle (PAM) robot arm as to improve its performance inmodeling and identification. The contact force variations and nonlinear coupling effects of both joints ofthe 2-axes PAM robot arm are modeled thoroughly through the novel dynamic inverse neural MIMONARX model exploiting experimental input-output training data. For the first time, the dynamic neuralinverse MIMO NARX Model of the 2-axes PAM robot arm has been investigated. The results show thatthis proposed dynamic intelligent model trained by Back Propagation learning algorithm yields both ofgood performance and accuracy. The novel dynamic neural MIMO NARX model proves efficient formodeling and identification not only the 2-axes PAM robot arm but also other nonlinear dynamicsystems.