Automatic Landing Control System Design Using Adaptive Neural Network and Its Hardware Realization

Abstract

This paper presents an adaptive neural network, designed to improve the performance of conventional automatic landing systems (ALS). Real-time learning was applied to train the neural network using the gradient-descent of an error function to adaptively update weights. Adaptive learning rates were obtained through the analysis of Lyapunov stability to guarantee the convergence of learning. In addition, we applied a DSP controller using the VisSim/TI C2000 Rapid Prototyper to develop an embedded control system and establish on-line real-time control. Simulations show that the proposed control scheme has superior performance to conventional ALS under conditions of wind disturbance of up to 75 ft/s.