Hi, this is Wayne again with a topic “Drones Take Flight: Andreas Raptopoulos, Matternet”.
Good morning um, thank you for having me here. I have a story around drones as well, and it’s interesting. It fits into crazy story a little bit i’ll refer to it in a minute um. We started this comical matter net.
We want to create a transportation network for matter the name tells the mission of the company. We want to create a network for transportation of matter that is inspired by the ideas of the internet and it’s a very young idea. Still it started off a couple of years ago just down the road from here at singularity university as nasa ames, and this is a place that you know. People come together in a summer program and they try to figure out ways to solve humanities. Grand challenges and the challenge that we focused on was extreme poverty and pretty soon we realized that there is a very strong correlation between people living in extreme poverty and transportation networks around them. So today, there’s one billion people in the world that do not have access to all season roads. Imagine this one billion people right, one seventh of the year’s population are totally disconnected for some part of the year. We cannot get medicine to them. They cannot get critical goods and they cannot get their goods to market in order to create a sustainable income and come out of poverty. In sub-saharan africa, for instance, 85 percent of roads are unusable in the wet season. Investments are being made, but at the current level it’s estimated it’s going to take them 50 years to catch up and catch up to what in the us, we have four million miles of road or more very expensive, to build very expensive.
To maintain. Is this the future that would like to see in you know the virgin continents of this world? So we saw this and we thought you know, there’s got to be a better way. Can we figure out a system? Can we create a new system using today’s most advanced technologies that can help these countries leapfrog hold this thought for a minute, let’s switch over to the other part of the world. You know the future of this world is mega cities. We live in cities. In you know, more than half of these population today lives in cities.
More than half a billion of us live in mega cities. These are cities, we thought more than 10 million inhabitants in them. What they all share in common is congestion. We haven’t been able to figure out ways to scale infrastructure, road infrastructure fast enough to serve those booming economies.
These are just some stats for the us. You can imagine the situation in you know many of the emerging world. You know: brazil, india, china, these stats will be much less favorable there. So when we saw that we thought you know, it’s not just a question, can we leapfrog we have to leapfrog, we have to figure out a way to lift some of that transportation.
Roads load off the roads and we think that’s the answer. We want to be to basically create a network of transportation using small uavs. When we started off this project, we thought, let’s just you know, go all the way the killer. App of you know. Transportation is human transportation, let’s just try and figure out a way to. Basically, you know strap a set of sensors that the small uav uses on a go-cart and have it fly.
You know at nasa ames we thought in two three weeks. We can pull it off and we thought okay, what’s the best place to start, we logged on to 3d robotics website water. We ordered an arduino kit, a small quad and we started sort of playing with it and um. You know the idea was to basically get this thing. You know together and running, fly it around and then try and figure out.
You know how can we scale it up uh for it to serve like a better, bigger payload? Now, as we started looking at, what could we do today at this small scale and those small sort of quads or octos or hexes? The name refers to the amount of motors that they carry. They can transfer today say two kilograms. We started thinking, you know: what can we do with those two kilograms, and this is the idea.
I think that really gelled in our minds and we created something that we think could be the next paradigm shift initially for lightweight transportation. So we want to create a network of small uavs, each one covering a small range about 10 kilometers today and transporting a small payload about 2 kilograms, but between them have a network that allows a relay type system that allows them to cover a big area and Over time, through repeated um, repeatably executing reliably transportation links be able to uh transport big loads by aggregating, you know at the point of destination, so this is what we call the matter net. We have three key technologies.
The first is small uavs. The second is those um trusted locations on the ground. We call them ground stations we swap in and out of them in order to exchange batteries and go out again or swap loads, and the third is the routing software.
This ai inspired engine that runs the whole network. It matches supply to demand it roots vehicles where they need to be in order to avoid weather conditions and other risk factors and so forth. Let’S look at these three components in some more detail. The videos we use today we use mostly of the shelf stuff.
Some of it is from chris’s platform, some others we’re experimenting with. We want to basically figure out a way to make transportation using small, uavs, very, very reliable and very, very safe, so we use an optocopter today. It’S able to transport two kilograms over 10 kilometers at a speed that allows us to cover those 10 kilometers in 15 minutes. This type of transportation capability may seem small for a real application, but if you put it in a city environment with you know, 10 kilometers you can span, you can go anywhere in the city in 15. Minutes. Compare this to you know, sitting for hours in traffic jams, you know in sao paulo, the heart of our system. Is this automatic battery and load swap mechanism? So the idea is that we land on the ground station and we swap a battery and load automatically at the beginning, because most of our missions are going to be in the developing world. This is going to be manually done by you, know human labor.
Over time. We try to figure out ways to do this automatically, so we have a very reliable router for our system. The ground station is a place where we can land and have a safe environment to do this battery and load swap, and this can take different forms. This is like the simplest form. It only is is, if you compare it like with the internet analogy again, we have things that sit in the background and do the routing and ground stations that are, like you know, front user facing sort of terminals such as this one, where someone can basically go And get or they you know, pick up or put in a package in the system drone comes in takes it goes out again. Uh we’d like to see a future of uh.
You know of transportation of uh last mile. You know. We think that this could be the solution to solving the last mile delivery problem, which is a big problem, that a lot of people are trying to figure out in all parts of the world and maternity os. This ai system that basically resolves how the network is operating for us, the main jobs that it does is that it does vehicle routing. It does package routing and it manages the whole network. You can imagine a situation where you have say you know a network like this.
Two thousand vehicles, 2 000 stations serving a location like you know, downtown sao paulo. You have bad weather coming from the east. What do you do? How do you make decisions? How do you route the vehicles? How do you make decisions when to land part of the network? How do you shut off part of the network for security, all that sort of stuff um, the you know on the on the front end of this. We want to basically maintain an image of whatever is flying in the national airspace, in whichever jurisdiction and be able to show these transparently to the authorities, and they should be able to have access to this. And you know, if there’s any sorts of emergencies you know be able to ground them down and so forth. One of the key challenges here, of course, is regulation in most places in the world.
You cannot operate this type of network today, autonomous flights are not permitted. So we’re taking a little bit of an activist approach and we’re trying to figure out places where you know there. The need is so high today that people would like to basically, you know, make exceptions to the rules or give a specific specific certificate of authorization.
For us to be able to run the networks, and i’m going to show you this from, can we run the video? Ah, you have the pdf, okay, all right so um. I was going to show you a video from one of our flights in haiti last summer. Uh, unfortunately, it’s not contained in the presentation for some reason, so we went out there and we basically tried to figure out.
You know. Where can we use this thing and we started basically prototyping uh different sort of usage scenarios and getting people around us to help us understand? How can we use this, and how can you know these guys be early adopters, uh? One of the crucial questions is, you know: what’s the cost today, and that depends a lot on the assumptions that you make on you know: what are the base costs for the different components of the system? What is their lifetime so forth? Today we have a cost of about 24 cents for transporting two kilograms over 10 kilometers, the energy cost of in that is only two cents this. What is this is, what gives us hope that this could be a new paradigm, so two cents to transport 10, kilometer at 2 kilograms over 10 kilometers in its cl. It’S clean energy, it’s very efficient and you know most in most um legacy – environments, at least very, very quick and, of course, we’d like to see the cost going down and reliability going up again.
We want to leverage all the great work that is happening in you know many places in the world now not just in open source communities, but in many universities and so forth. I think we’re going to see amazing things happening in this drone space. Little drone space and we’re going to be able to take advantage of this. So you know we have all sorts of things we’re looking at humanitarian applications.
Query applications like i’ve talked about before we’d like to see you know those little stations in rooftop, swapping batteries and creating small transportation networks, and the idea here is to basically create the next type of network in the world. That’S a network for the transportation of mother. Thank you.
.