The Drones and Aerial Robotics Conference happened last weekend in NYC. I was really looking forward to going but an impending deadline at my day job kept me in LA. I wasn't able to watch the livestream, but I was able to experience the conference through twitter; primarily via the #droneconf tag. I made a list of favorite tweets that resonated with my interests and ideas of what's important in drones. Ingrid Burrington/@lifewinning, Allison Burtch/@irl, Adam Rothstein/@Interdome and Ron Evans/@deadprogram especially did some good twittering.
Some people think all drones are evil. Other people think those people are making an error by conflating military drones and hobbyist UAVs. I don't think all drones are evil, but the magnitude of the societal change that both civilian and military/law enforcement drones will bring is so big that we need to discuss and debate both. Changing the name won't change the need for that debate.
Vijay Kumar is director of the GRASP Lab at UPenn, where they program multicopters to perform impressive aerobatic feats and fly in precise swarms. They do an amazing job attacking the drone programming challenges he listed.
Regarding "Drones are missing operating systems to allow people to develop apps", the earliest prototypes of that are coming, but wouldn't it be great if there was a Processing for drones? The open source drones are finally moving past the weak micro-controller stage, but it feels like it's still going to be years before they reach the level that the AR.Drone has set, where you can start getting creative and run interesting code on the drone: 128 MB RAM and running Linux. (And any tight control loop, like visual tracking, needs to run on the drone itself.)
Possibly the most interesting thing about DARC is that it brought together people creating drone technology and people thinking about the social effects of the technology. Knowing the technology intimately and working on it first hand is exactly why I am both excited for and scared of what the future holds. When the About page of this blog says "You now live in an age of ubiquitous flying robots. ... Cheap flying robots have already changed the world, but it took a while for people to notice" I'm hoping it communicates the combination of hope and concern that I feel.
The legal situation regarding drones in the U.S. is a mess. Drone startups are blooming here, but they must either fly in other countries or live in fear of an FAA shutdown.
The FAA has grounded the University of Nebraska-Lincoln's Drone Journalism Lab and the University of Missouri's Drone Journalism Program. It has put a stop to many attempts to use drones in search and rescue situations, and also stopped local police forces from flying drones. At the Small Unmanned Systems Business Expo a few months ago, I even heard Ted Wierzbanowski (former Industry Co-Chair of the FAA’s Small UAS Aviation Rulemaking Committee) say that he thought if a farmer bought an RC plane and flew it over his land to collect aerial imagery to see how the crops were doing that it would not be legal under current FAA rules.
At the moment the only way for public agencies or companies to fly drones without getting in trouble with the FAA is to get a Certificate of Authorization or a Special Airworthiness Certificate (though if you look at this list of companies with certificates, it appears you have to be a defense contractor to get one). It will be a big step forward if the FAA really issues new guidelines in the next couple months, but I'm also nervous about the possibility of new rules that are even more restrictive than the ones we have now, especially for hobbyist activities.
Almost everything gets hacked. And with unencrypted hard drives, unencrypted video downlinks, Windows-based control stations, and maybe even unencrypted GPS it's like the military isn't even trying. There is good news, though: DARPA is working on secure drone software, and one of the most popular open source drone communication protocols may soon have a secure version.
Essam Attia is the artist who was arrested for putting up NYPD drone posters. I didn't know he'd worked as a military geospatial analyst!
When I saw Daniel Suarez talk about drones at DorkbotSF last year I was all like OMG this guy sees the same crazy drone future that I see and am kind of freaked out about! His book Kill Decision, a techno-thriller about killer autonomous drones unleashed by rogue elements of the military-industrial complex, isn't superb literature but extrapolates from current technology to capabilities that are both plausible and shocking.
Web of Drones
The workshop I was going to do at DARC (before I flaked) was going to be about the Internet of things + drones. This was the proposal I wrote on the fly, literally at the last minute:
The future of drones will be as egalitarian (and as risky) as the web.
That is part of the future I see for drones.
Missy Cummings is Director of the MIT Humans and Automation Lab and was one of the first female fighter pilots in the U.S. Navy. In the tweets above she touches on two of my favorite themes: The idea that eventually robots will do every physical activity (and some mental activities) better than we do, and the human factors of drone control.
The only document I've seen on the human factors of drone ground control stations is the "UAS Ground Control Station Human-Machine Interface Development and Standardization Guide", published by the office of the Under Secretary of Defense for Acquisition, Technology and Logistics, but I've been collecting visual catalogs of DIY GCSes and military/commercial GCSes (as well as one of Primary Flight Display/Electronic Flight Systems).
My Mavelous project is intended to be an extremely easy to use GCS that can run on smart phones and tablets (Missy Cummings has worked on simple iPhone interfaces as well), but I've also spent a significant part of my career making natural language interfaces to autonomous systems—the idea is to reduce operator workload by talking to your drone like you can talk to Siri (see my "Electric Familiar" post for a video).