Unmanned Aerial Vehicles Sensor-Based Detection Systems Using Machine Learning Algorithms

Abstract

Detecting Unmanned Aerial Vehicles (UAVs), also known as drones, is becoming more difficult as technologies keep advancing. The low price, smaller size, and high speed of UAVs make them hard to detect. The goal of this study is to critically review and evaluate the UAVs sensor-based detection systems using Machine Learning (ML) algorithms. The study reviews several sensor-based detection systems (acoustic, thermal infra-red, radio frequency, and radar), and makes recommendations for future enhancements using machine learning-based techniques. One of the findings of this study is the small amount of data used by researchers, due to the lack of publicly available datasets, which added limitations to the research and may have produced inaccurate results. Another important finding is the closed environments (labs) that most researchers have conducted their research in, which are far from real case scenarios. Finally, this research makes recommendations on how to improve the process and obtain more accurate results. Classification and identification of UAVs are beyond the scope of this paper.