Real-Time Helicopter Flight Control: Modelling and Control by Linearization and Neural Networks

In this report we determine the dynamic model of a miniature helicopter in hovering flight. Identification procedures for the nonlinear terms are also described. The model is then used to design several linearized control laws and a neural network controller. The controllers were then flight tested on a miniature helicopter flight control test bed the details of which are also presented in this report. Experimental performance of the linearized and neural network controllers are discussed. It was found from experimentation that the adaptive neural network technique was able to improve helicopter flight performance