Automatic Landing System Using Adaptive Resource Allocating Network

摘要

This paper presents an adaptive resource allocating network (ARAN) deign to improve the performance of conventional automatic landing system (ALS) and guide the aircraft to a safe landing. Real-time learning is applied to train the ARAN that uses gradient-descent of the error function with respect to the weights to perform the weights updates. Adaptive learning rates are obtained through the Lyapunov stability analysis. Convergence of learning is guaranteed. Simulations show that the proposed scheme has better performance than the conventional ALS.