Development of autonomous flight control systems for unmanned helicopter by use of neural networks

This paper describes approaches to develop command and control systems for unmanned aerial vehicles (UAVs). YAMAHA RMAX, which is an unmanned helicopter, is used in this study. The dynamics of RMAX is nonlinear, so that it is hard to develop autonomous flight control systems, but an efficient method to design controllers by training neural networks is proposed in this paper. Methods to develop controllers for feedback linearization and robust control systems are described and numerical simulations show the effectiveness of our method.