Adaptive trajectory control for the twin rotor MIMO system using artificial neural networks

Abstract

This paper investigates the development of an adaptive trajectory dynamic non linear model inversion control law for a twin rotor multi-input multi-output system (TRMS) utilizing artificial neural networks. The behaviour of the TRMS in certain aspects resembles that of a helicopter. A highly non linear one degree freedom model of the TRMS is considered in this study and a non linear inverse model is developed for the pitch channel. In the absence of the model inversion errors, an artificial neural network (ANN) model in place of a proportional-integral-derivative (PID) controller is used to enhance the tracking performance of the system. The neural network model is developed using backpropagation algorithm with Levenberg-Marquardt (LM) training method. The responses between the reference signals and empirical based models of the TRMS to validate the accuracy of the models. Simulation results under MATLAB/Simulink prove the improvement of response and superiority of simplified neural network controller.