USING A NEURAL NETWORK TO IMPROVE THE COURSE STABILITY OF UAV IN THE GNSS DENIED ENVIRONMENT

Abstract

This paper is devoted to the development of a concept for improving the accuracy of keeping the course of airplanetype unmanned aerial vehicles in the absence of satellite navigation signals in conditions of unknown wind load using
neural networks. The paper describes the results of the development and training of a neural network based on data
collected during UAV training flights. The results of the work can be used to modify existing UAV flight controllers in
order to improve the accuracy of their autonomous functioning in the conditions of electronic warfare.