Monday, June 4, 2012

After months of development and heartache, the journalism drone takes flight

Months of planning, training, re-design, and fabrication finally paid off for developers, as we successfully launched a journalism drone for the first time on Sunday, June 3, in the small Illinois town of Tuscola.

Eventually the drone will be equipped with an Arduino-based autopilot (APM 2.0) and cameras to collect aerial photography and aerial photomaps for use in journalism investigations. JournoDrone 2 is the successor to JournoDrone One, which was meant to be a "Unmanned Aerial System (sUAS) for journalists that is powerful, durable, transportable, affordable, upgradeable and supported by a community of experts."

JournoDrone One was smashed to pieces during testing in two months ago. Since then, I had been applying the lessons learned from the first drone to make a drone for journalism that was stronger and more capable.

The June 3 flight was a basic shakedown of the airframe, without autopilot or sensors, to prove the airworthiness of the drone. While the airframe proved slightly heavy and unpredictable at times, our indication is with some modification it will be a competent flyer for our purposes.

JournoDrone 2 is based on the "FPV Raptor" radio controlled airplane with some important modifications, the most important of which is carbon fiber-reinforced fuselage. The airplane had a fairly good track record as a First Person Video drone, so we didn't feel that we needed to change the layout of the craft. Yet we needed to have an extra layer of security to make it more of a "tool" rather than "toy."

While we had some initial concerns about weight gain, the two layers of carbon fiber and epoxy proved its worth when the aircraft experienced a sudden gust of wind in its second flight and crashed. While the impact separated the motor pod, chipped the propeller, bent the motor shaft, and cracked the casing of an on-board camera, the fuselage was totally unscathed. We were not able to recover the on-board video of that second flight.

I owe a debt of gratitude to University of Illinois physics professor Mats Selen, a fantastic teacher and RC expert who not only helped immensely with the setup, balancing, and repair of JournoDrone 2, but also provided his piloting expertise for the first flight. He is pictured in the right on the topmost photo.

Mats and I, along with fellow developer and UIUC GSLIS PhD student Acton Gorton, are using this learning experience to help our Drones for Schools initiative.

Through National Science Foundation (NSF) Math Science Partnership (MSP) grant, EnLiST, we hope to better STEM education by giving high school students the opportunity to build and program these drones in the classroom and use them to provide aerial intelligence in their communities.

More repair and development needs to go into JournDrone 2 before we can outfit it with an autopilot and sensors. The motor pod needs rebuilt and strengthened, and some of the wing needs repaired as well. The electronics need to be secured better in the fuselage and the center of gravity needs to be balanced appropriately. And finally, the battery needs to be tested to see if it is still safe to be charged (LiPo batteries damaged by drops or shocks may burst into flames when charged).

For now, here are some photos and a video from the flight that day in Tuscola:

Selen (left) and Gorton (right) making final preparations to JournoDrone 2 before the second flight.
Selen and myself (right) Testing radio equipment before flight in Tuscola.
Selen and JournoDrone 2, which had a camera attached for its final flight of the day. The camera shut off on impact, however, deleting all of its footage. A problem with the electronics shifting in the fuselage affecting the center of gravity could have contributed to the crash on this flight.

And last, a video of the first flight, piloted by Dr.Selen and recorded by Gorton.