UK biomimetic engineering startup Animal Dynamics is building a microdrone with wings inspired by the flapping flight of a dragonfly. The project, which started in June 2015 with a feasibility study, is being funded with £1.5 million from the UK Ministry of Defence, via DSTL, the Defence Science and Technology Lab.
Last fall the company switched from researching the feasibility of the concept into phase two: actually trying to build the thing. They now say they’re confident they’ll have a flying prototype of their Skeeter drone to demo by this summer — with the tech potentially deployed in the field by the end of next year.
To be clear this microdrone has not yet got off the ground. At this point all they’re showing publicly is the bug-like model on a stick, pictured above. But Animal Dynamics co-founder and CEO Alex Caccia says he’s confident it will take to the air in “two to three months’ time”. “One of the challenges with something that flies is that everything has to work for it to work at all — but we’re pretty close to it now. The ‘ah-ha’ moment of it flying is almost the last thing that happens,” he adds.
The team is also looking to raise around £4 million in Series A funding in the next few months for continued development of Skeeter but also to fund some potential spin-out technologies they’ve created along the way — such as a high efficiency linear actuator designed for the drone which they reckon could also be used for other use-cases, such as in medical pumps and for road vehicle propulsion.
“We’re fundamentally interested in developing commercial products from studies and understanding of how nature reaches these tricks that allow greater performance and efficiency,” says Caccia.
Drones with flapping wings do exist — including a DARPA-backed drone that resembles a hummingbird, built by US company Aeroenviroment back in 2010 — but it’s fair to say that flapping drones are the exception not the rule. A far more typical animal in this space is the buzzing quadcopter.
Yet rotary blades have drawbacks. They don’t support stable flight in windy conditions. They’re noisy. They can even be dangerous. And they can require a lot of power to stay airborne. Hence the MoD’s hope of driving development of a more robust flight technique that can withstand tough in-the-field environmental conditions.
“It’s a very extreme challenge, set down from them — DSTL came up with the requirements — which is can you make something at this scale operate in high wind and difficult environmental conditions,” says Caccia, discussing the MoD’s requirements for the Skeeter drone.
“They’d been using small drones in Afghanistan and Iraq with quite a lot of success because when the environmental conditions are right they are extremely useful for soldiers on the ground to go out and see what’s round the corner… they need to be small so that they can’t be seen, so that they’re easily carried and so that they’re quiet. However as soon as there’s a slight wind — anything above 5 meters per second — they get blown out of the sky… So they’ve got a frustration.”
As Caccia tells it, the ‘usual suspect’ defence suppliers weren’t at all confident they could build anything to meet the MoD’s microdrone challenge. But Animal Dynamics’ other co-founder, Adrian Thomas, a professor of biomechanics in the Zoology Department of Oxford, suggested the answer could lie in looking to nature for inspiration — given that birds and insects are able to achieve stable flight in turbulent conditions. And, ultimately, Animal Dynamics’ pitch secured the DSTL funds.
“Adrian was doing some work in his garden during Storm Doris and there were 50 mile an hour winds and there were bumblebees happily buzzing around the lawns, completely unfussed by the high winds. Which is something insects have solved for a very long time,” says Caccia. “It’s very, very difficult to do but the interesting things is that flapping wing propulsion lends itself to solving this problem very well. Rotary blade propulsion doesn’t.”
Asked for an opinion on the engineering challenges of flapping, Dr Mirko Kovac, director of the Aerial Robotics Lab at Imperial College London, tells us: “Flapping wing flight has several advantages compared to propeller based solutions, including the ability for high manoeuvrability and potentially low energy consumption during forward flight. The challenge however is significant and includes the need for a thorough understanding of the aerodynamics involved, as well as the development of the mechanics and wing transmission mechanisms as well as the controller for successful flight.”
Commenting on the Skeeter project specifically, Kovac adds: “The mentioned timeline seems possible but it will depend on the size and weight of the vehicle. Bird-sized flapping wing vehicles are partially already available on the toy-market while bee-sized flying robots are still the topic of intense university research. However, I do believe that it is possible to build a flapping micro drone that can provide value in environmental monitoring, smart farming and search and rescue applications.”
Caccia says the biggest remaining challenge to getting Skeeter off the ground at this point is the mechanical design. Flapping, as you’d expect, is a lot more complex in engineering terms than spinning — especially if you also have relatively little power to play with, as it’s a lightweight, battery-operated device.
“The challenge is really around producing a very low friction mechanism. So the wings we’ve built, the flight control system has been solved, it’s actually the mechnical design that’s very difficult. So we’re doing some work with some people in the Swiss watch industry to help us out… It’s really about friction. You need to get the mechanism to be very, very low friction,” he says.
“Even the slightest friction will cause resistance, and create heat and stop the thing from working properly. Most mechanical systems get around it by putting an unreasonable amount of power in. We don’t have that so we have to make things run very, very smoothly.”
Thomas also points to “friction and inertial load” as the hard problems. “The high leverage at the wing hinge means that the motors ‘see’ about 50 times the wing weight, so driving the wing weight down has huge benefits. Similarly, apart from the aerodynamic loads, almost all the work done by the motors is work against friction in the flapping system, driving the friction in the system down pays huge dividends,” he says.
“The stability and control systems may seem challenging, but there has been a lot of work done on control and stability in birds and insects, and our vehicle has a huge advantage over any of the other current drones that can hover — turn the motor off and it glides, with good passive stability, down to a relatively gentle landing.”
On the plus side, the team says it’s benefiting from the easy availability of electronic components — spilling over from the mobile industry.
“The extraordinary thing and one of the factors that has made a project like this possible is availability of components from the mobile phone industry. The access to very low costs MEMS and sensors and tiny antenna and a whole array of electronics is really one of the key factors,” says Caccia. “I’ve noticed that a whole bunch of components have come onto the markets as development boards — literally in the last two to three years — I think partly also driven by the wearables market… Which anyone can have access to. And certainly we’re using that.”
He also suggests this liberal availability of electronic components is providing an added incentive for the MoD to fund projects such as Skeeter. “They need to try and keep one step ahead, but also engage with the tech developer world far more to understand how these technologies are being used,” he adds.
The dragonfly-esque Skeeter is planned to be 120mm at its largest; weigh less than 20 grams (packing a camera and the other necessary comms and navigation sensors); and have a top speed of around 45km per hour. In terms of flight time Caccia says it will be “useful” — qualifying that as “not quite an hour” — though he also notes it depends on wind conditions, and points out that a drone with wings can also glide — thereby saving on battery power.
Range will be most limited by the radio signal which he says might be up to 1,000 meters. While the per unit price they’re aiming for is the “low thousands” — so the microdrone can be “widely used and in effect almost thrown away” — though it remains to see if they can keep costs down.
One thing is certain: should Skeeter get off the ground, this is going to be a very bespoke animal indeed — a drone made to measure for its military masters. But, while you’re in no danger of receiving an airfreighted Amazon package to your doorstep conveyed via an industrious team of Skeeter dragonflies, Caccia does reckon flapping wing tech holds promise for more than just stealthy surveillance microdrones. Especially as the form factor need not be so small. And it’s certainly true that military-funded technology has a habit of filtering down to the consumer space after the expensive R&D work is done — as indeed is the case with drone tech itself.
“I think there’s a market for it not just in the military but also elsewhere too, and also at different scales. There’s been a very clear focus requirement to make it at this scale, because they’ve been using something at this scale… but the technology can be scaled up. So one of the reasons we’re looking to raise funding is we’d like to make a bigger one,” he says.
“There’s all sorts of advantages you can have with a larger, flapping drone. Far, far more efficient flying from A to B. Can still hover. Much less dangerous. You can put your finger in the flapping wings as they flap and it won’t hurt you… And also with a quadcopter drone, if any of the mechanism fails it falls out of the sky like a brick. Whereas the things that we’re making glide in their neutral position.”
Could a larger flapping wing drone be capable of taking payloads — say for a delivery use-case? Caccia reckons it could, though he says it would need to have half a meter to a meter wingspan. Which, sadly, suggests there’s also little prospect of urban drone delivery via giant dragonflies. Maybe just for some edge cases — such as delivering humanitarian aid to remoter areas.
“I think delivery drones is a laughable idea,” adds Caccia. “I don’t think it’s really going to happen. It’s a sort of fantasy. But I think there are uses [for a larger-scale Skeeter] — for instance agriculture, for instance surveying large field areas.
“A quadcopter type thing is pretty nigh useless for that because the flight time’s so low. So I think it’s something that’s definitely worth exploring.”
Even at the microdrone scale, the team sees potential agricultural use-cases for flapping propulsion. “One area I’d like to explore is precision agriculture inside greenhouses,” says Thomas. “Using the drones to deliver precise tiny doses of nutrients or pesticides to the plants that need them rather than dosing the whole greenhouse, that might be a good use for the existing drones once we have them in mass production and have the cost down to sensible numbers.”
It is also investigating the potential of flapping in water, for propulsion and hydropower-generation. And is on its third prototype of a human-powered boat, also animal-inspired of course, with Caccia pointing out that fish swim with far greater efficiency than propeller-based water crafts. He describes the craft as looking “kind of like a recumbent bicycle with a dolphin fin behind it”.
“This is the activity that Adrian and I first got excited about,” he tells TechCrunch. “It’s something we’re hoping to be able to get out this summer — and have a go at breaking the world speed record. Just as a demonstrator of how you can make something flapping go very fast.”
“Propeller design efficiency has basically reached an asymptote, there’s been no real, material improvement in the efficiency of propellers in the last 20 years,” Caccia adds. “Of course everyone thinks flapping is a completely ridiculous thing to do but nature’s way of telling you you’re wasting energy in water is a stream of bubbles. And fish don’t produce a stream of bubbles when they’re going about… So we’re interested in all sorts of areas. We’re making an out-board motor that uses it. It’s also very, very quiet… And we’re making our human-powered boat. And it would be wonderful to see larger vessels using it too.”