Drones that can 3d-print...
Interview with
Mirko Kovac and a team of scientists at Imperial College London have been designing a type of flying drone, bio-inspired by swiftlets - tiny birds that build nests using their own spit. He brought in one of his Micro Aerial Vehicles, MAVs for short, to show Kat Arney what it's capable of...
Mirko - Drones are basically flying robots or flying unmanned aerial vehicles. So, what they can do is they can take pictures from above, from the sky, but it can also interact with the environment and create structures, repair structures, or take samples of the soil or of water.
Kat - We mostly hear about drones in the context of military spying or doing nasty things like dropping bombs and stuff like that. That's presumably not what you're about here.
Mirko - Of course, there are a lot of ethical questions and discussions that happen around these areas. What's different from these big military drones is that they are basically airplanes. But the small drones that we can buy for a few pounds are fuelled by consumer electronics and cell phone technology. So, they are really cheap now and they are widely available.
Kat - Now, I'm kind of a bit worried about the idea of lots of drones flying around that could take pictures of me. Would they take pictures of me while I'm sunbathing in my bikini? I don't really want that. How do I know what's a good drone or maybe a pervy drone?
Mirko - Yeah. So I think we need to discuss this as a society, what is ethical use of drones? There are different questions that come to mind. One is on privacy i.e., not to film people when they don't want to be filmed. I think we need to protect that. But we also need to take care that drones don't do something that is potentially dangerous. That they're not in the hands of terrorists for example or that they're not carrying out tasks that are critical, or dangerous, or making life and death decisions. I think there are things that the human needs to be in the loop and we need to create a framework, an ethical framework of what should be allowed and what's not.
Kat - So, we've got a couple sitting on the desk here. Talk to me about this one. This thing is I guess the size of the span of my hand. It's got 4 little helicopter blades sitting around something about the size of an egg I guess. What's this here?
Mirko - So this, I would call it a nano quadcopter. It's very small and they can carry some sensors. We use those in research to swarm based control and swarm coordinated actions. We can have thousands of them that will be deployed from a bigger airplane or from a mother ship drone, and then create and interact together to collaborate and do things.
Kat - Like a flock of these tiny things going out there.
Mirko - Exactly. So, they could be in a flock that can build structures like termite woods in flocks, or even swarms collectively, or they can also just sample environments because we have many of them. Even if we lose part of them, even if we lose 80% let's say, they're very cheap and disposable. So, we still get a lot of data back.
Kat - Okay, let's put this one to the test. Shall we do a little test flight here in the studio?
Mirko - Let's do it. I can try.
Kat - Off you go. Okay, so that seems to be slightly hard to control. How are you and your team working on how to control these little drones and take them out in the world?
Mirko - Yeah, I'm not a very good pilot for these drones. So, that's why we work on autonomous control so they can control themselves based on on-board sensing and on-board intelligence. This is one of the key areas that we work on. What we also work on is on the platform design. So, creating new drones that can do new things such as repairing structures.
Kat - So, how would that work? How could something that's a small little flying drone build a structure? How does that work?
Mirko - For that, we look at nature and we look at the animal kingdom in particular, these birds that use their saliva to build nests. So these birds basically invented 3D printing many million years ago and so, we copy their principle of additive layer manufacturing to build up structures. Now, we do this with expanding foam on the drone.
Kat - Not spit.
Mirko - No.
Kat - Unlike the birds.
Mirko - No, the drones is a polyurethane foam or could use any kind of other chemicals that can layer up and just build up structures or repair structures.
Kat - So, I can see that could be really useful for very far away environments or possibly really dangerous environments. I guess we heard about the Fukushima nuclear reactors. You wouldn't want to send people there to repair it, would you?
Mirko - Yeah, that's exactly right. It's a very good application. There are areas where people don't want to go, but also importantly, there are areas or situations where quick response is required such as an oil spill or a leak in an oil pipeline. So, they're very important to quickly react to this, a drone could go and seal this leak very quickly and effectively.
Kat - We've also heard recently in the news about the idea of delivering things with drones or online shopping or something like that. Perhaps even a pizza. I'm feeling quite peckish. Where are we with that?
Mirko - I think we're pretty close to that. So, we're not so far, but there are a lot of challenges on how to fly drones safely in cities close to humans, how to avoid each other, how to have a regulatory framework that allows the regulation of the traffic, like air traffic control. But the technology is almost there for delivery.
Kat - What could be some of the applications? I mean for example we're, in the grip of the Ebola crisis at the moment, one big problem is you can't get medications, you can't get tests through. Is this the kind of area where drones could really help?
Mirko - I completely believe so. In fact, in areas where the infrastructure is not developed yet such as developing countries, so instead of going into building roads or building ground infrastructure, there's a big movement in creating aerial drone based delivery systems at a fraction of the cost that would be much faster and more effective for medical deliveries, blood supplies for example.
Kat - Obviously, the need in some parts of the world is incredibly urgent for this, but how soon do you think we might start seeing little drone armies or drone swarms going out and doing these kind of things?
Mirko - So, I think the drone delivery is very close, so we need to create legislation for that. We need air traffic control, but we also need on-board sensors and ways how to avoid each other, how to avoid obstacles. Then again, I think we can learn a lot from nature that does it already very, very effectively.
- Previous Driverless cars
- Next Predicting solar eruptions
Comments
Add a comment