T4: Drones as Edge Devices: Challenges, Technologies, and Applications

Organizers and Presenters

YangQuan Chen
Director, Mechatronics, Embedded Systems and Automation Lab (MESA Lab)
Professor, Dept. of Mechanical Engineering
School of Engineering, University of California, Merced, CA, USA

Kimon P. Valavanis
Director, Research and Innovation
D. F. Ritchie School of Engineering and Computer Science
University of Denver, CO, USA

Derek Hollenbeck
Mechatronics, Embedded Systems and Automation Lab (MESA Lab)
Dept. of Mechanical Engineering
School of Engineering, University of California, Merced, CA, USA

Tutorial Summary

During the Drones: Technology, Policy and Society workshop organized jointly by CITRIS and Honeywell at UC Berkeley on Oct. 25, 2018, one of the invited panelists and speakers Brad Westphal, senior director of Honeywell, introduced the company’s end-to-end inspection and data analytics services using small UAVs. While the program primarily serves aircraft, renewables and oil and gas companies, Westphal addressed the scale needed to improve safety and lower costs across commercial industries. “Within each one of these businesses, there are opportunities for drones and UAVs to provide efficiency and extract value,” he says. In his talk, he suggested drones are being used as edge devices. A full recorded talk can be accessed from this link: UAV experts present the latest in drone tech.

This has sparked a new wave of development regarding drones as edge devices. Edge computing devices are everywhere with considerable computing power that can be employed to run complex algorithms to perform different applications. As the edge computing power increases, Artificial Intelligence (AI) algorithms can be embedded in the edge devices. With Edge AI, introducing Real-Time data analytics to a drone application is possible without depending on a centralized server to perform the corresponding analysis and visualization tasks. Thus, drones can be used in industry settings not only as moving sensors for inspection, but also as local intelligent data processor and prognostic tool. Edge AI for UAVs/UASs opens new possibilities to expand the processing capabilities on industry settings, contributing to growing Industry 4.0 automation, analysis, and data-based process.

Topics to be Covered

  • YangQuan Chen: An Overview of Edge Computing, Industrial Artificial Intelligence (IAI) and a Fault Face Deep Learning Case Study
  • YangQuan Chen: Process Industry Use of Drones as Edge Devices
  • Kimon Valavanis: Comparative Studies of Model-based and Data-Driven Techniques and Complexity Analysis for UAS
  • Kimon Valavanis: Flying Quadrotors in Confined Environments for Infrastructure Inspection: Experimental Validation and Verification
  • Derek Hollenbeck: A Digital Twin Framework on Using Drones as Edge Devices for Environmental Monitoring 
  • Derek Hollenbeck: Methane Sensing Drones with Machine Learning
  • YangQuan Chen: ET (evapotranspiration) Sensing Drones with Machine Learning
  • All: The Road Ahead

Intended Audience

This Tutorial is suitable for graduate students, researchers, scientists and engineers, practitioners, end-users and developers interested in autonomous UAVs. The collective outcome of the Tutorial is an understanding of what may be achieved when autonomous UAVs are treated or regarded as edge devices with edge AI embedded.

Tutorial material

Participants will receive detailed presentations and papers.


  1. The Edge AI, Artificial intelligence and edge computing for smarter controls, by MESALab