Oh little drone, how you have grown…

Article By : Brian Dipert

Drones are something of a design nightmare: the batteries needed for long flight times add weight that shortens flight time, and on-board intelligence causes added charge drain.

I wasn’t planning on buying a drone, but as long-time readers have likely already figured out, I also find it hard to resist a compelling tech deal. As such, when I saw a factory-refurbished first-generation DJI Mavic Air on sale for $369.99 at BuyDig back in late May, I took the bait—hook, line, and sinker.

Its color scheme is Arctic White; I admittedly would have preferred Onyx Black (but not Flame Red, the third option offered with this model):

Then again, it’s $430 off the original $799 MSRP, so I’ll get over the non-optimal tint. My prior purchase reticence was based on two primary factors: drones’ reputed propensity to annoy neighbors, strangers, and pretty much anyone else around you when you’re flying one; and my limited but underwhelming previous experience with one.

More than a decade ago, iFixit’s Kyle Wiens and I had briefly played with his Parrot AR.Drone, a sibling of which his firm had recently torn down, at ESC Silicon Valley. Kyle was admittedly more adept than me—I couldn’t get the thing to stably fly, while he was zipping it around in the press room. Although the AR.Drone did include a single camera, it was only VGA in resolution and didn’t support recording. This trendsetting quadrotor was also big and bulky, and it could only best-case fly for around 12 minutes per battery charge (with recharge taking ~1.5 hours). And as for object avoidance or any other autonomy (or even assistance)…fahgettaboudit.

How far we’ve come in less than a decade (the first-generation Mavic Air was introduced in January 2018):



It’s not like I’d completely ignored the topic of drones; after all, I’d done teardowns of three generations’ worth of “toy” versions within the last half-decade-ish:

And I’d even published a “top 10” list in mid-2016.

The seeds of my purchase to-come, I realize in retrospect, were probably planted when I stumbled across the Wilderness Sessions series of videos published by Holly Bowling in late 2020. Like me, Holly’s a Deadhead (we’re also both Phish Phans). Unlike me, she’s an accomplished pianist. And when the world entered pandemic lockdown, she and two travel companions (one human, one canine) jumped in a camper van and traveled across the United States, periodically stopping to video-record performances in various locations. Her music is stunning (IMHO, at least); equally mind-blowing is the drone (and stationary) video footage captured by her husband, Jeffrey. Salt Flats, Utah may be my favorite:



But the entire 12-video sequence, including the “behind the scenes” retrospective, is well worth a viewing. Just reserve plenty of time, because once you start the series, it’ll likely be hard to stop until you get to the end.

One convenient thing I didn’t realize until post-acquisition is that, unlike its less expensive “Mini” series counterparts, the Mavic Air supports autonomous object avoidance (of trees, walls, etc.), thanks to six additional cameras at front, rear and underside (one pair each), coupled with onboard processing intelligence. It also has IR sensors underneath for added “intel” of the ground below it, and two IMUs (inertial measurement units) to assess angle versus the horizon and other position, orientation, speed and acceleration metrics.

Put together, this all means that (unless you’ve disabled some or all of these “training wheels”), the Mavic Air is pretty much impossible to crash. And it also includes an onboard GPS receiver. This means that not only can it log coordinates as it flies, it knows where it started…and since it has a pretty good idea of how much battery life is left, once it reaches an appropriate turnaround point, it’ll automatically head for home. So it’s pretty much impossible to lose, too. Reiterating what I said earlier, we’ve come a long way.

On that point, the successor Air 2, introduced in April 2020, and Air 2S, unveiled in April 2021, offer even more:



They make predictable improvements in various areas, such as:

  • Object avoidance (the Air 2S adds upward-looking cameras, for example)
  • Main camera performance (resolution, video frame rate, low-light image quality, etc.)
  • Additional autonomous flying and filming modes, and
  • Longer per-battery flight time

On that last one, however, if you look closely at the comparative descriptions I link to earlier you’ll quickly notice that the newer models are quite a bit bigger and heavier than the first-generation Mavic Air is. The bulk of that added bulk (ha!) constitutes bigger batteries.

Li-ion battery technology is pretty mature at this point, as is electric motor technology, so in the absence of a fundamental high-volume technology breakthrough in the future, to get longer flight time, you need to include bigger batteries…which leads to what I find most fundamentally fascinating about drones and their flying kin: the fundamental balancing act of trading off various contending design factors that is unique to the craft of engineering (versus, for example, pure R&D or science). Look at what I’ve just said. Everyone wants to be able to fly their drone as long as possible, before needing to land and swap out battery packs. But in order to do so, that means that the drone manufacturer needs to include larger battery cells, and more of them.

Added bulk admittedly has the side benefit of making the drone more tolerant of wind gusts, for example; but fundamentally, the heavier the drone the beefier the motors need to be in order to lift it off the ground and fly it for meaningful altitudes, distances, and durations. Beefier motors burn more juice, which begs for more batteries, which make the drone even heavier…see the quagmire? And unlike with earth-tethered electricity-powered devices, you can’t just “pull over to the side of the road” if the batteries die on you.

Now toss in the added “twist” that everyone also wants their drone to be as intelligent as possible so it doesn’t end up lost or tangled in branches, and so it can automatically follow whatever’s being videoed. All those image and other sensors, along with the intelligence (and memory, and..) to process the data coming off them, burns juice, too. And don’t forget about the wireless connectivity between the drone and the user—minimally used for remote control and analytics feedback to the user, and in more modern drones (not mine) also for FPV (first-person view) real-time video streaming of camera images to goggles on the user’s head:



And used to capture incredible one-take footage like this:



And this:



How do you balance all of those contending factors to come up with an optimum implementation for your target market? For one knowledgeable “take” on the answer to this question, I’ll close with a great interview my boss in my “day job” (editor-in-chief of the Edge AI and Vision Alliance) recently did with Hayk Martiros, head of autonomy at Skydio (another leading drone manufacturer). The public preview version is below; for the whole thing, head to the Alliance website.



And with that, it’s time to wrap up. Admittedly, I haven’t even taken the Mavic Air out of its box, so I’m sure I’ll have more to say in future posts. Until then, I welcome your thoughts in the comments!

This article was originally published on EDN.

Brian Dipert is Editor-in-Chief of the Edge AI and Vision Alliance, and a Senior Analyst at BDTI and Editor-in-Chief of InsideDSP, the company’s online newsletter.

Leave a comment