This episode of Real Engineering is brought
to you by Brilliant, a problem solving website that teaches you to think like an engineer. In the past few years there has been a lot
of buzz around the possibility of drone delivery services. Most think companies like Amazon will be the
first to deploy them in their distribution centres, but few realize the technology is
already in use for a far greater purpose in the developing world. More than two billion people across the world
lack access to essential medical products, like blood and vaccines, due to poor quality
or even non-existent infrastructure. Last month I visited Rwanda with Sam from
Wendover Production and Joseph from Real Life Lore. We were all struck by how modern and clean
Kigali, the nation's capital, was but just a short drive outside the city the quality
of roads deteriorated quickly. The country has been working hard to improve
their transportation links over the past 2 decades, but of the 14,000 kms of roads in
Rwanda, only 2,600 kilometres are paved. The rest consist of uneven dirt roads that
become incredibly difficult, if not impossible to navigate during the countries raining season. [1] We got a taste of this while on Safari when
our 4x4 got stuck in the mud, forcing us to get out and push, keep in mind we were looking
for lions on this trip. Medical supplies, by nature, need to be on
hand quickly. If a mother is bleeding out during childbirth,
she cannot afford to wait 3 hours for blood to arrive. Leaving people living in remote rural villages
in danger. Zipline, the real reason we visited Rwanda,
is working to solve this problem with their fleet of autonomous drones. Each capable of carrying 1.6 kilograms of
medical supplies, about the weight of a three 500 millilitre blood bags. These amazing little drones have a TONNE of
geeky engineering design that was influenced by Zipline’s design philosophies. Several unique factors influenced their designs. The first is the speed the plane needs to
get into the air. The moment an order comes in, the clock has
started and Zipline aims to have the drone in the air in as short a time as possible. This is, after all, an emergency and seconds
count. Currently they are averaging just 5 minutes
from order to launch. That’s the time it takes for an order to
arrive in their on-site pharmacy to launch. Getting a plane into the air that quickly
is pretty astounding, some would struggle just to get a drone out of its case and flying
in that time, and Zipline have come up with some amazing solutions to reduce those seconds. Just like a DJI Drone, Zipline’s plane needs
a GPS connection in order to fly as they are not piloted, they are autonomous. If you have ever flown a drone you know it
can sometimes take a little while for the drone to boot up and make a connection with
a GPS satellite. So, to remove that delay, Zipline moved the
GPS circuitry from the plane to the battery. This means it’s always on and always connected. This innovation alone removed on average 10-15
minutes off the launch time. The battery is one of 4 pieces of the plane
that need to be assembled prior to flight. First the order is placed inside the drop
hatch of the fuselage, which is then placed on the launcher, the wings are then attached
followed by the battery. This modular design makes the plane much easier
to handle, allowing staff to easily lift it into place. More importantly, it separates components
so if there is a problem found with the wings during the pre-flight check, they are simply
swapped out without having to start the entire assembly process over again. Pre-flight checks are often a lengthy process
and Zipline have come up with some cool solutions to hasten this step too. Checking the flight surfaces is done with
a mobile app that connects to the launcher system. The launch technicians simply point the phones
camera at QR codes on each control surface, sending a message to the plane to actuate
the control surface. The phone then utilizes a computer vision
algorithm to make a pass or fail judgement for each control surface. Once the plane is ready and assembled on the
launcher, the next time saving measure kicks into gear. Rather than telling you how this works, I’m
just going to show you. -Cut to launch footage.- We spent 2 days at Zipline and this never
got old. The rail uses a pulley and an electric motor
to quickly and safely get the drone up to speed. Accelerating the plane to its 100 kilometre
per hour cruising speed in just 0.3 seconds. It launches the same way every time and reliably
clears the obstacles in front of it. Take-off and landing are the most difficult
stages of a flight. To avoid having to land at the destination
the plane simply drops the supplies in a insulated cardboard box with a simply parachute, which
can be thrown away. Meaning the clinics need no infrastructure
to sign up as a client of the distribution centre. When the plane does eventually need to land,
the procedure to allow it to land with no pilot is even more incredible. Once again, I think it’s easier to just
show you how it works. Cut to footage -
It may be difficult to keep track of everything that happened there, so here is that again
in slow motion. There is a small hook at the end of the tail
boom, which will grab a wire strung between two actuated arms on either side of the capture
system. It’s hard to see it in real time, but in
slo-motion to can clearly see that the arms actually raise up at the last moment to catch
the hook. The plane is communicating its location in
3D space with the radio receivers next to the platform, allowing the wire to stay out
of the way until the last moment. Minimising any risk of collision with the
wire. The capture system misses about 10% of the
time, but the plane is programmed to immediately detect a miss and throttle up it’s engines
to gain altitude and make another pass. An older design of this system, had a retractable
tail hook that caught an actuated wire closer to the ground, which then slowed the plane
down enough for it to landed softly on an inflatable pad. This system was fraught with design issues. When it rained the inflatable would pool with
water, which was no problem for the water proof plane, but forced the staff to crawl
through it and lift a relatively heavy plane. It was neither comfortable or ergonomic. The retractable tail boom needed a motor to
control the retraction, which could fail and added weight compared to the simple metal
hook attached to the carbon fibre tail boom of the current generation plane. Factoring safety into an autonomous drone
network is another unique factor that has shaped Zipline’s designs. Zipline works on a philosophy of redundancy
of parts in order to ensure safety. It has two of everything. Two motors, when it only uses one during flight. Two of every actuator, and if somehow two
of something breaks the plane will automatically deploy a parachute. Allowing it to softly touch back down to earth. This procedure can also be initiated by the
control tower if a command from the air space authorities is received. As a result of this focus on safety Zipline
has had zero accidents causing injury since they started service in October 2016. The next design factor, which will be common
to any aeronautical design. Is the weight of the aircraft and in turn
it’s range. Zipline uses planes because quadcopters use
far more energy to fly, and just can’t get the range necessary from batteries and are
relatively slow. Map Animation:
These planes cruise at 100 km/h with a range of 160 kms, allowing the drone to serve an
80 km zone around this location. The inner skeleton consists nearly entirely
of a light weight carbon fibre composites, which is then covered in a light weight foam
shell which is easy and cheap to replace if it gets damaged. As a result, the entire fuselage weighs just
6.4 kilograms. The wings weigh 2 kilograms with wing span
of 3 metres. The structurally integral wing spar, that
is the beam that runs along the length of a wing, is also made from a high strength
carbon fibre composite. With the aerodynamic surfaces being formed
with high density polystyrene, and 3D printed plastics. The batteries are by far the heaviest part
of the vehicle, making up half of the total weight of the aircraft at 10 kilograms. Zipline employed lithium ions batteries with
a total capacity of 1.25 kwhs. For comparison, a Nissan Leaf has 24 kWhs
and your average Tesla has about 120. As mentioned before, these removable sections
aren’t just batteries. They contain the GPS circuitry, but also include
the data storage that hold the flight data from all the sensors on board. The moment the battery is hooked up to charge
at this charging station, the data begins to poor into Zipline’s server. For every hour of flight these drones generate
1 gigabyte of data. This for me, is Zipline’s greatest asset. These drones don’t just face engineering
challenges, but regulatory challenges too. Fitting a large autonomous network of drones
into the already busy and highly regulated American airspace would be incredibly difficult. Rwanda serves as not just a worthy cause,
but a valuable test bed for integrating a network of autonomous drones into a countries
air traffic control. Zipline communicates directly with Rwanda’s
central air traffic control in the Kigali Airport, in the nation’s capital. Having this test bed, in my mind, is more
valuable than any hardware technology Zipline has developed. Ryan Oksenhorn, one of Ziplines founders,
gave us a guided tour when we arrived. He’s Zipline’s head of software and has
worked tirelessly to make Zipline’s autonomous drone control system the best in existence. A quick look through patents attributed to
Ryan will uncover a slew of designs related to automated drone management systems, like
Patent 9997080 “Decentralized air traffic management system for unmanned aerial vehicles”,
which describes software solutions to allow multiple UAVs, which have no way of detecting
other UAVs in their airspace, to avoid collisions. This testbed and data is going to allow Zipline
to continue to churn out designs and concepts to optimize the control of the drone delivery
network, and ultimately allow them to expand into new territories with busier airspaces. This month they are expanding within Rwanda
opening a second location to serve the east of Rwanda. This could eventually grow into a supply chain,
allowing drones to hop between bases, get their battery swapped out in a couple of minutes,
and be on their way once again. This technology is useful outside of just
developing countries. It could be used in disaster relief scenarios
when roads become impassable due to flooding. It also allows supplies to be centralized. Hospitals often have to carry more medical
supplies than they need. Keeping stocks of medical supplies with short
shelf lives inevitably leads to waste. Incredibly expensive waste that costs the
taxpayer money. This is a problem Zipline could potentially
reduce by allowing centralization of an emergency supplies that can quickly be dispatched when
needed. When quarantines are an issue, these drones
could minimize human exposure to contagious diseases, and so help stop the spread of disease. These are both scenarios that Wendover Productions
and Real Life Lore explore in their videos released today. We could not have made this trip without the
help of Brilliant, so if you would like to see more videos like this please check them
out. If you have an interest in this kind of design,
I would highly recommend you take this course on classical mechanics, which will form the
bedrock of understanding for you to start applying physics, like figuring how much batteries
to include for a drone design. This is just one of many courses on Brilliant,
with more courses due to released soon on things like automotive engineering and Python
Coding. If I have inspired you and you want to educate
yourself, then go to brilliant.org/RealEngineering and sign up for free.And the first 73 people
that go to that link will get 20% off the annual Premium subscription. As always thanks for watching and thank you
to all my Patreon supporters. If you would like to see more from me the
links to my instagram, twitter, subreddit and discord server are below.
Ooo my company helped with designed the launch and catching mechanism. So cool to see a video posted about it
Rwanda is a really interesting place in general. It's incredible if you go there how clean/organized/safe it is. I've been to 7 countries on the continent and it was by far. Wouldn't think it was the site of the genocide in the 90's, except for the fact that there's not a lot of older folks.
Edit: I even stayed in THE hotel Rwanda, very cool.
Really excellent video on a rather clever system. IIRC, though, it's operated in Rwanda by Rwandans but was designed by a Californian engineering company.
Impressive engineering.
This is an absolutely beautiful project - I really hope it can gain some more traction in other countries if it proves useful.
The whole system is so magnificently simple and spartan. If you try to do this in the EU, the cost could be a hundred times higher. Now, i understand that someone might not appreciate biohazard samples flying over their head on a dumb drone with no fancy failsafes and duplicates, but dammit why can't we do things the simple way sometimes?
This is so cool.
This is so amazingly cool.
This is a beautiful simple system. Anyone have any idea why they have to do a flight check before launch? I mean, I know you have to do that. But I'm wondering why they don't have a plane ready that you can just load and go.
Also this makes me wonder about a gas version with much longer range. It could be a great way to deliver stuff to remote places in Canada.