Постановка наукового завдання удосконалення систем навігації та управління безпілотних літальних апаратів у дестабілізуючих умовах

Abstract

The paper analyzes methods for autonomous UAV navigation and control in destabilizing conditions, such as GNSS-denied environments and electronic countermeasures. The impact of non-Gaussian noise and outliers on inertial system accuracy is examined. The study proposes integrating Visual SLAM with Euclidean Signed Distance Fields (ESDF) for safe trajectory planning. Reliability is enhanced through robust filtering (H∞ -MCUKF and AMEECKF) to suppress interference and stabilize state estimation. The prospects of using AI, including reinforcement learning and recurrent neural networks, for adaptive control and obstacle avoidance under sensor data deficiency are substantiated.