Cleo Robotics Demonstrates Uniquely Clever Ducted Fan Drone

This doughnut-shaped drone, not technically known as a dronut, offers a tasty combination of safety and ease of use

4 min read

Evan Ackerman is IEEE Spectrum’s robotics editor.

Cleo Drone
Cleo drone prototype at CES 2017.
Photo: Evan Ackerman/IEEE Spectrum

At last year’s CES, Cleo Robotics was showing prototypes of a palm-size drone with a design unlike anything we’d ever seen. Shaped like a doughnut, the Cleo drone is essentially a ducted fan, with a pair of completely enclosed propellers (one on top of the other) and then a camera, battery, and electronics housed inside the shell. It’s compact (95 millimeters in diameter, 33-mm thick, 90 grams), elegant, and inherently safe, since the nasty spinny bits are all tucked away. With fewer motors than conventional quadrotors, it promises to be more efficient as well, and quite possibly cheaper. But if you look closely at the picture, you’ll probably end up with the same question that I did:

How the heck does the Cleo drone steer?

It’s easy to see how changing the relative speeds of the propellers could get the robot to yaw, but pitch and roll were a mystery to me. Back at CES, Cleo wouldn’t tell me how it was done, but assured me that their drone was fully controllable. My best guess at the time was that there was some sort of system of movable weights inside the drone, sliding around to unbalance it in the direction you wanted it to go.

Now Cleo Robotics has just released a new video of an operational prototype of the Cleo drone, and Cleo’s founder and CEO, Omar Eleryan, was kind enough to provide us with (a few) more details about how it works.

Here’s the video of Cleo in action:

It’s immediately obvious just how friendly this design is. It fits in a pocket without needing to be disassembled or folded, and there’s nothing externally fragile that you have to protect. It’s safe to hold, and so grabbable that you can snatch it right out of the air. Collisions with obstacles (including people) shouldn’t damage either the drone or whatever it runs into, and with all of the moving parts so well protected, it seems like it has to be much more durable than anything with exposed rotors.

While it’s true that there are a few other drones out there that offer this level of protection (like the Flyability Elios and the Hover Camera Passport), they’re relatively bulky, and you also pay a penalty in battery life. Cleo, meanwhile, boasts a useful battery life in the 12- to 15-minute range, which is excellent for a drone of its size, leveraging both design efficiency and the increased amount of thrust that a ducted fan generates.

Centralized ducted fans have been used in drones before, but without any exceptions that we’re aware of (besides perhaps this one), all of those drones use big vanes underneath the fans as control surfaces, steering themselves by deflecting the airflow. You can see that Cleo has no big control surfaces underneath, and is essentially symmetrical at the top and bottom. There’s a (very) limited amount of detail that Cleo Robotics is willing to publicly divulge about their drone’s control system at this point, because as far as they know (and the PCT patent office backs them up on this), it’s completely novel and unique. I can say for sure that it doesn’t involve movable weights (oops), but here’s the extent of what Omar Eleryan was able to tell us on the record:

“We introduce control surfaces into the airstream to change the direction of the airflow and create a thrust vectoring effect.”

We wish we could tell you more, but based on this information and what you can see from the pictures and video, there are some assumptions you might be able to make about where those “control surfaces” are and how they might work. 

Cleo DronePhoto: Cleo Robotics

Eleryan tells us that Cleo Robotics was founded after an experience that anyone who has ever bought a drone can almost certainly relate to: 

This all started when I bought my first drone, and my co-founder Simon and I took to a park. We set it up, and instantly the excitement that we had to fly the drone turned into anxiety. The drone was big, it was loud, and there were kids running around, and we were scared about what might happen. Ten minutes later, we got it stuck in a tree. And we ended up wondering how we went from being so excited about this awesome technology to being so frustrated with it.

There are three problems that we were trying to overcome [when we founded Cleo Robotics]. One of them is practicality; when we first started out drones were really big, and for the most part they still are. They’re also quite intimidating, because they have exposed propellers. For most people, they’re not comfortable having a drone indoors, or around their kids, for that reason. And the last problem is that drones are very difficult to fly. 

The idea behind Cleo was to create something very easy to use, very practical, and very safe. Our goal is to create a safe and practical drone that we can seamlessly integrate into the workplace and daily life. It’s not just about creating a thing that flies—it’s also about creating something that will have a real use, and solve a real problem.

Of course, the recent history of drones is littered with the remains of startups like Zano and Lily that promised innovative designs but failed to deliver. Cleo Robotics acknowledges that it still has work to do. The company has had a working prototype of their drone for some time now, so their recent focus has been on giving it some intelligence and autonomy. The prototype that you see in the pictures and video just has one camera, but Eleryan says that the final version will use a combination of stereo-vision-based SLAM and neural networks for indoor navigation and obstacle avoidance.  

In terms of applications, Cleo is thinking about how its drone might be useful in primarily indoor environments, for security and surveillance, for example, rather than as a consumer camera drone (at least initially). This will likely come as a bit of a disappointment to those of us who were hoping to be able to buy one of these, but Eleryan says they are not giving up on a consumer version: After all, that was a big reason why he started Cleo in the first place. At the same time, though, it’s refreshing to see a small drone startup targeting something besides glorified selfies, and actually trying to make a drone that’s useful and practical in other ways.

The company says you can expect to see the Cleo drone come to the security market within the next 12 months or so, and they’re hoping to have a consumer drone that’s priced competitively with other small quadcopters by the second half of 2019.

[ Cleo Robotics ]

The Conversation (0)