DroneTrack: Cloud-Based Real-Time Object Tracking Using Unmanned Aerial Vehicles Over the Internet

Abstract

Low-cost drones are emerging on the horizon of novel smart Internet-of-Things (IoT) applications. Researchers in cloud robotics have recently pushed for the integration of low-cost robots and drones in the cloud and IoT. However, the performance of real-time cloud robotics systems remains a fundamental challenge that demands further investigation. In this paper, we present DroneTrack, a real-time object tracking system involving a drone that follows a moving object over the Internet. DroneTrack uses the Dronemap Planner (DP) cloud-based system to control, manage, and communicate with drones over the Internet. The main contributions of this work consist of: (i.) the development and deployment of DroneTrack, a real-time object tracking application using the DP cloud platform, and (ii.) a comprehensive experimental evaluation of its real-time performance. We note that DroneTrack does not use computer vision techniques; rather, it relies on the exchange of GPS locations through the cloud. Three scenarios are presented for conducting various experiments with real and simulated drones. A tracking accuracy of 3.5 meters on average is achieved by DroneTrack with slow-speed moving targets. Our experimental study demonstrates the effectiveness of the DroneTrack system.