High Speed Object Tracking Using Edge Computing

Joel Dick, Caleb Phillips, Seyed Hossein Mortazavi, Eyal de Lara

Poster at the 2nd ACM/IEEE Symposium on Edge Computing (SEC), San Jose, CA, October 2017



The use of unmanned aerial vehicles (UAV), or drones, has in recent years seen explosive growth due to lower costs and technology ad- vances in mobile computing, batteries, sensors, and control systems. Drones are now used in a multitude of applications, from natural resource exploration, the film and entertainment industry, to ur- ban surveillance, and defense. The image processing demands of these applications requires higher powered computing capabilities than those available locally to the drone, prompting the offloading of these tasks to the cloud. However, the latency requirements of the cloud are beyond those acceptable for many applications. This paper proposed the use of a server on the network edge to opti- mize both processing capability as well as latency for applications requiring real-time communication between a drone and a cloud server. We propose to test the limits of this model by implementing a system for real-time tracking of golf drives on a golf course.