People clearly liked the idea: The team raised well over $1 million on Kickstarter to produce its faithful flying friend. After 14 months of development, delays and a few awards, the product is finally shipping, beating GoPro's Karma to the skies. But, while AirDog's trick might be to follow its owner, it's already in the lead -- in more ways than one.
I tried an AirDog prototype more than a year ago, and Engadget got another look more recently as well. But it was only last month that I finally got a peek at the version you can actually buy. I've flown a number of drones, but AirDog is different in a few key ways. It's simpler to use (pretty much push a button and go), and it's designed to fit an active lifestyle (more on this later). Just days after getting my hands on the AirDog, GoPro CEO Nick Woodman would list these exact two qualities as being core to his product, Karma, during an interview at CES. Interesting!
What do I mean by "fit an active lifestyle"? Example: AirDog uses a waterproof wrist-worn gadget called the "leash" (geddit?) as both its controller and GPS tracker. It's what tells the quadcopter where you are. Other drones aimed at the action-sport crowd use your phone's GPS -- perhaps not ideal for snow- or water-based rough-and-tumble. Also, unlike most rival follow-me drones, AirDog uses downward Lidar sensors to stop it hitting the ground, should you suddenly descend too fast. Meaning, if you take a steep drop on a mountain bike or ski run, AirDog won't bounce off the rocks, only to follow you downhill in broken pieces.
Lidar is OK for preventing downward collisions, but avoiding things like trees and buildings is still an issue you need to consider. True obstacle avoidance for consumer drones is only just starting to mature, so you're still only a rogue branch away from a crash if you haven't planned your route well. This is one area where GoPro might actually have an advantage, if the company's recent acquisitions are anything to go by.
Perhaps what sets AirDog apart from current rivals are the little details -- details that should have the team at GoPro paying attention. The most obvious example is the fold-out design. Those purple limbs collapse into the body making AirDog incredibly rugged and portable. If you remember the original Decepticon Transformer "Soundwave," AirDog reminds me of one of his "cassettes": a slimline slab of angular plastic that opens up into a flying robot (it's basically Laserbeak).
Most of the other big-name camera drones I've flown require a case about the size of hand luggage (once you've factored in all the accessories) and still feel delicate in transit. Interestingly, I have heard from reliable sources that GoPro's Karma is also going to be foldable and compact and will work with existing GoPro cameras and not be a camera in and of itself (as others have reported).
One tradeoff of AirDog's "flat" design is that it has a more limited gimbal (the thingy that keeps the camera steady) than most quadcopters. Usually, the camera hangs underneath, giving it the ability to rotate left or right (or "yaw"). AirDog has the camera up front, and, much like the urban legend that our canine friends can't look up, AirDog can't look side to side (without moving the drone).
Here's where things get interesting: The gimbal might lack pivot range, but it has far greater potential, including access to deep features within the GoPro camera. Most companies can't access a GoPro's internal API (a set of tools that provide access to key features on the camera). The first product to be allowed access was also a drone: 3DR's Solo. Both Solo and AirDog connect to the GoPro via the larger port on the rear and not the USB on the side (like DJI's Zenmuse and other gimbals do), and this is the port you need to get enhanced control of the camera.
A few people in the industry have described use of this port to me as "the Holy Grail" of GoPro access. Right now, AirDog doesn't have any extra control of the GoPro; it currently just powers the camera. But it appears to be designed with this in mind. At least technically it could, and it's safe to assume the team is pushing for it.
AirDog has confirmed to Engadget that it does indeed have access to GoPro's API/settings, use of which will be rolling out in a future update to add features that make use of it.
Here's the thing: AirDog the company is a tiny, dedicated team. It doesn't have the financial sway to compete with GoPro, so why would the camera company even consider handing over access to a rival product? Perhaps the same reason it gave access to 3D Robotics? It's speculation, but one of the unique features of 3DR's Solo are the cable-camera modes, which help create exactly the same effect you see in GoPro's latest Karma-shot video (video linked to the exact moment). It wouldn't be unreasonable to suggest GoPro is leveraging access to its API in exchange for knowledge.
Woodman even hinted at CES that "more so than any other drone, [Karma] will be designed to be woven into somebody's active lifestyle." (Emphasis ours). Of course, he would say this, but couldn't you be even more confident about that if you'd been keeping your biggest rivals on a short leash? GoPro reportedly already tried courting DJI, with CEO Frank Wang claiming the camera firm wanted a larger slice of the profits than he felt fair.
AirDog folds down to a compact size, perfect for backpacks.
What else does AirDog have to offer? I'd say an intimate knowledge of exactly the kind user GoPro is going after. Just using the AirDog makes that obvious. Unfold the legs, and it auto-switches on and locks onto GPS/connects to the leash. Press one button and it's up and flying, ready to follow.
A simplicity, again, eerily in line with Woodman's rhetoric about his own company making products that are easy to use, like the one-button Hero4 Session. AirDog also has different follow modes (straight line like a cable, or fluid following, for example). You can also ask it to circle you. All of these features are accessed via the leash, a waterproof box smaller than a pack of cigarettes. If anything, controlling AirDog is much easier than operating a GoPro with its similar-looking wireless remote.
The more I explore AirDog, the more apparent the conclusion seems to be: It's not just the action-sport drone to beat, it's a team GoPro should consider buying. If AirDog is the high-water mark, GoPro knows only too well it has to do better. What easier way to eliminate that threat than to swallow it whole?