Unmanned aerial vehicles using machine learning for autonomous flight; state-of-the-art

Abstract

In recent years, since researchers began to study on Unmanned Aerial Vehicles (UAVs), UAVs have been integrated into today's everyday life, including civilian area and military area. Many researchers have tried to make use of UAVs as an ideal platform for inspection, delivery, surveillance, and so on. In particular, machine learning has been applied to UAVs for autonomous flight that enables UAVs do designated task more efficiently. In this paper, we review the history and the classification of machine learning, and discuss the state-of-the-art machine learning that has been applied to UAVs for autonomous flight. We provide control strategies including parameter tuning, adaptive control for uncertain environment, and real-time path planning, and object recognition that have been described in the literature.