As drones get smarter, so must their owners

On Sunday, April 17th, British Airways flight BA727 from Geneva to London was coming in for a landing when the pilot saw something -- reportedly a drone -- strike the front of the aircraft. The collision was said to have happened above Richmond Park, an area where drone flying is prohibited (but not because of its proximity to the airport). Perhaps more worrying was that the impact happened at 1,700 feet -- well above the 400-foot height limit for all drone flights in the UK, even in permissible areas. Fortunately, BA727 landed safely and without incident. It was also probably a false alarm. But with more and more amateur drone pilots taking to the skies, how long until it happens for real?

If the BA incident had involved a drone, I might have had mixed feelings. Last summer, I was testing a drone. To evaluate the device and its camera, I wanted to find a safe yet picturesque place to fly. I chose a quiet stretch of beach, a short bus ride south of where I live in Spain. I say quiet: That was until two police officers rolled over the sand dunes on quad bikes and told me to stop. It turns out that I, too, was in a national park. How can I roll my eyes when I've done the same thing? (I've never flown at 1,700 feet, though -- nor will I ever.) In my (weak) defense, I knew it was near a national park, but thought I was outside the boundary.

Almost a year later, I'm testing another drone: the FlyPro XEagle Sport. This time, I want zero chance of encountering the police (or airliners full of people). But it's no simple task. In fact, for a car-less city dweller, finding somewhere scenic and legal without plenty of planning is hard. But tough noogies, right? As more drones fly themselves, filming us autonomously, we need to be more careful. A smart drone still needs a smart human.

The XEagle is one such drone -- a "follow me" quadcopter for capturing sporting moments, similar to AirDog. The "Sport" edition has no ground station (controller). Instead, the XEagle follows a wearable that looks like an '80s-era Casio for kids. You can buy a ground station for it, but all I have is the wrist-watch thing. All the more reason to find somewhere open and away from people.

With no "sticks," you have less immediate control over a drone. The XEagle lands at the press of a button, and there are dials on the watch to move it up/down and left/right, but this is still hardly what I'd call "control." Even then there are problems. If there's a barking dog or an object below it, or it's now unsafe to land where you took off, you don't have many options. The main photo for this article is the XEagle landing in a pothole it obviously couldn't see. In fairness, it righted itself and returned to the spot where it took off; it just did a little gardening along the way.

When searching for a good place to fly, Google Maps is an obvious place to start. I used its measurement tool to draw a five-mile line from the nearest point of the airport (the distance Spain's aviation authority requires). I then drew a circle with that radius in Photoshop. This gave me an easy view of the most important no-fly zone where I am (Valencia). Local laws also require you fly in unpopulated areas. That means nowhere residential, obviously.

This effectively means I must leave the city, as there are no parks big enough to comfortably avoid people. Valencia's airport is to the west, meaning I must focus my search on the east -- which is hemmed off by the sea. Beaches are off-limits if there are people, so it's either gamble on an early-morning flight or look elsewhere. Given my experience last time, I decide to look elsewhere.

You might be thinking, aren't there apps that make it easy to see no fly zones? There are. There's the FAA's B4UFLY and Air Map, among others. The problem is, knowing where you can't fly isn't the same as knowing where you can.

There are tools for this, too. But none seems to have nailed it. Drone Zones lets users add spots, but lacks a large community. DJI's +Discover has the potential user base, but is relatively new, and flying hot spots are only a minor feature within the app. DJI compiles data from users' flights to map these "hot spots" automatically. At least it does in China. The version for Europe and the US did offer this feature, but privacy concerns led the company to put the feature on hold. When I asked DJI why it removed hot spots, a spokesperson said: "While we initially launched the hotspot feature outside China, we weren't satisfied with the experience it created and temporarily removed that functionality."

For the brief moment it was there, though, I found two hot spots near me. One within my "airport circle" (it's since disappeared), so that's out. The second was near a town called Alboraia just outside the city. Google Earth shows the spot to be wasteland surrounded by fields. It's not going to be scenic, but for testing the XEagle, it could be ideal.

The patch of land is a short metro ride from where I live, tucked away behind the town. It's the sort of place where people dump old furniture and fridges (I determine this from the old furniture and fridges lying around). It's otherwise barren -- perfect for testing a drone that may or may not follow me around, as advertised.

The reason DJI doesn't want to include flight hot spots in the new edition of app is that there's not enough data to safely anonymize things. If I saw a DJI pilot at this piece of wasteland, for example, there's a statistically high chance it'd be the person from the app. That said, the feature's success in China shows what can happen when the location data are used wisely. If there were an app that offered the same thing for all drones, it'd be the perfect discovery tool. As you can see from the screenshot above (Shenzhen left; Valencia right) good use of data is the key to finding safe, scenic places to fly.

Of course, there's a broader, more serious point to be made here. I'm just flying for fun and photos. But many of the same problems I face apply to more serious applications, as Drone Advocate Peter Sachs reminds us. "Unlike traditional model aircraft, drones are useful," he says. "To be useful, drones will (and should) be used in urban areas." Sachs points to journalists or first responders who are using drones in the city for good. "If populated areas were to be considered off limits, it pretty much defeats the whole purpose of them using drones."

In that context, I'm suddenly happier with my little patch of dirt. And I'm pleased that technology helped me find it. It's perfect for testing the XEagle. I've learned that the drone follows you well, but it's prone to random adjustments. It offers decent battery life (about 20 minutes' flying time) and while it comes with a camera, you'll probably want to upgrade it to a GoPro (the included one is a cheap clone). When it goes on sale for $489, it will make for a decent introduction to autonomous flying, though it lacks the polish of its pricier rival, AirDog.

More importantly, I enjoy the peace of mind of knowing I can legally fly it. That's especially true as I continue to master the drone's quirky controls. For example, rotating the dial on the watch-thingy's face moves it around you in a fixed circle. But, in practice, that only works if you're moving (I guess so it knows which way is "front"). Occasionally the drone realizes it's behind me when it should be in front. Dutifully it flies over me to correct that. This kind of behavior would be perilous if there were trees or power lines around.

And that's the thing. With autonomous flying comes an uneasy feeling. Perhaps it's healthy nerves. Perhaps it's a reaction to relinquishing control. Perhaps it's unreasonable fear that a "BA727 incident" could happen to even the most well-intended drone pilot. As apps like Drone Zones and DJI +Discover show, the smarts to solve these problems are already there; it's up to the drone community to apply them. They don't just apply to urban fliers, either. Everyone wants to find that literal sweet spot. Don't just tell us where we can't fly: Help us find the places we should fly.