A Review on Deep Learning Techniques for Unmanned Aerial Vehicles Navigation

Abstract

Over the past decade, there has been a significant advancement in the research and development of Unmanned Aerial Vehicles (UAVs) and Deep Learning (DL) methods, which, when used in conjunction with path planning algorithms, help UAVs achieve autonomy. UAVs have been flown autonomously using a variety of techniques, but modern deep learning techniques such as Convolutional Neural Networks (CNN), Imitation Learning (IL), and Deep Reinforcement Learning (DRL) are some of the techniques that have recently been favored. Different approaches like path planning, localization, and obstacle avoidance are used. Multiple approaches using CNN, RL, and IL were put out in this area. The majority of the literature on the subject since 2020 is reviewed in this study. The goal is to compare different methods used in the autonomous navigation of a UAV. Different existing algorithms and methods have been discussed and explored in the category of deep learning.