WEBVTT 00:00.000 --> 00:02.000 You 00:30.000 --> 00:32.000 You 01:00.000 --> 01:02.000 You 01:30.000 --> 01:32.000 You 02:00.000 --> 02:02.000 You 02:30.000 --> 02:32.000 You 03:00.000 --> 03:02.000 You 03:02.000 --> 03:04.000 You 03:30.000 --> 03:32.000 You 03:32.000 --> 03:34.000 You 03:34.000 --> 03:36.000 You 03:36.000 --> 03:38.000 You 03:38.000 --> 03:40.000 You 03:40.000 --> 03:46.000 You 04:00.000 --> 04:02.000 You 04:02.000 --> 04:04.000 You 04:04.000 --> 04:06.000 You 04:06.000 --> 04:08.000 You 04:08.000 --> 04:10.000 You 04:10.000 --> 04:12.000 You 04:12.000 --> 04:14.000 You 04:14.000 --> 04:16.000 You 04:16.000 --> 04:18.000 You 04:18.000 --> 04:20.000 You 04:20.000 --> 04:22.000 You 04:22.000 --> 04:24.000 You 04:24.000 --> 04:26.000 You 04:26.000 --> 04:28.000 You 04:28.000 --> 04:29.000 You 04:29.000 --> 04:31.000 You 04:31.000 --> 04:32.000 You 04:32.000 --> 04:33.000 You 04:33.000 --> 04:34.000 You 04:34.000 --> 04:35.000 You 04:35.000 --> 04:36.000 You 04:36.000 --> 04:37.000 You 04:37.000 --> 04:38.000 You 04:38.000 --> 04:39.000 You 04:39.000 --> 04:41.000 You 04:49.000 --> 04:50.000 You 04:50.000 --> 04:52.000 You 04:52.000 --> 05:22.000 So, what I mean is that, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean 05:22.000 --> 05:52.000 I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean, I mean 05:52.000 --> 06:02.280 well, hello everyone. 06:02.280 --> 06:06.180 Thank you so much for all being here. 06:06.180 --> 06:11.320 I really didn't expect this turnout at what Sunday morning, so yeah, thanks. 06:11.320 --> 06:21.120 We are Benny and Mark and we are developing affordable airborne wind energy systems 06:21.120 --> 06:32.120 So I already explained what actually that is in just a minute, but first let me start with our problem that we have. 06:32.120 --> 06:45.120 So the demand for renewable electricity is rising faster than we all know, but so far the expansion of the electricity production in renewable area is actually very incomplete. 06:45.120 --> 06:55.120 Because solar power has become so cheap and simple that everyone can now produce electricity, so does that mean that our problems are solved? 06:55.120 --> 07:00.120 Well, no, because what happens if the sun doesn't shine. 07:00.120 --> 07:11.120 And because as cheap as producing solar power is as expensive it is actually to store it so far, so that is why we need wind power. 07:12.120 --> 07:18.120 Unfortunately, wind power is very much lagging behind at least where we are from in Germany. 07:18.120 --> 07:40.120 Wind turbines are, if you think about it, the opposite of solar because they are, they need to reach high enough to reach the wind and so they need to be, yeah, as large as possible and that makes them resource intensive, expensive and that requires lengthy planning and also last but not least they are not universally. 07:40.120 --> 07:50.120 Loved for the appearance. So what if it was possible to make wind power as accessible as solar power? 07:50.120 --> 08:02.120 Everyone wind energy promises exactly that as it replaces the concrete tower structure and the foundations with a simple line. 08:03.120 --> 08:10.120 And we actually know this principle from steel robot kites. 08:10.120 --> 08:26.120 When it flies and figure aids or circles, you are pulled by the kite and everyone wind energy systems harvest this energy by replacing your hands with a spool, which is attached to a generator. 08:26.120 --> 08:37.120 Reaching higher where the winds are stronger, maybe even twice as strong as most wind turbines can reach nowadays is as simple as just buying a longer line. 08:37.120 --> 08:45.120 But making this simple concept work is actually not very easy. 08:45.120 --> 08:55.120 So far projects were way too complicated and thus expensive, like this project from Google, which is for that exact same reason that now. 08:55.120 --> 09:03.120 Let's get projects when investments start flowing in, set the industry back with all the advancement lusts. 09:03.120 --> 09:08.120 So, well, let's make this simpler, right? 09:08.120 --> 09:12.120 We realized for that we need to take a lean and open approach. 09:12.120 --> 09:19.120 We start small instead of big and what we can't handle and we make our advancements open. 09:19.120 --> 09:22.120 That secures them and allows participation. 09:22.120 --> 09:33.120 Five years ago, we started from scratch seeking to do as much as we can ourselves and based on open source technology, which is already there and doesn't cost anything. 09:33.120 --> 09:40.120 So, after many iterations and many flight tests, we can autonomously take off, generate power and land. 09:40.120 --> 09:52.120 And our system ended up with basically 10 times lower production costs than other projects and we achieved that through the lowest complexity by far. 09:52.120 --> 09:59.120 For example, we can take off vertically with only two propellers, we can land without propellers. 09:59.120 --> 10:05.120 Our ground station is exceptionally simple as you can see there. 10:05.120 --> 10:12.120 Additionally, our autopilot is also very simple and easy to understand and modify for the user as well. 10:12.120 --> 10:16.120 And it doesn't need any GPS to know where it is. 10:16.120 --> 10:24.120 And we also have very high safety standards, based from the concept of self actually. 10:24.120 --> 10:28.120 And we are for that reason compatible with at traffic. 10:28.120 --> 10:31.120 For example, we can implement at traffic surveillance. 10:31.120 --> 10:36.120 And we have the fastest landing by far so we can avoid obstacles. 10:36.120 --> 10:45.120 We have redundant control surfaces and we have a feature for an autonomous emergency detection and landing. 10:45.120 --> 10:52.120 Also, that's the most important part maybe is the highest flexibility of such a system because it is portable. 10:52.120 --> 11:00.120 It can be deployed in minutes and the flying area can still be used for any other purpose at any time basically. 11:00.120 --> 11:11.120 And of course for us here, the most important aspect is we don't secure ourselves any patterns and all electronics and software are open source. 11:11.120 --> 11:15.120 So, but how much power can we actually make with this? 11:15.120 --> 11:26.120 So, what you see here, this kite generates 1 kilowatt on average and 5 kilowatt peak. 11:26.120 --> 11:32.120 And that means the entire energy consumption of one German person including heating and driving. 11:32.120 --> 11:40.120 And if we enlarge the wingspan to 6 meters, we already get 5 kilowatt average and 25 kilowatt peak. 11:40.120 --> 11:45.120 And that equals the entire energy consumption of an average German farm. 11:45.120 --> 11:52.120 However, this enlargement would not make the production of this system very much more expensive for us. 11:52.120 --> 11:58.120 Hopefully, this would be quite sensible to do, which is why we're going to do that. 11:58.120 --> 12:04.120 And so, as you can see, the wingspan doubles, the output multiplies by 5. 12:04.120 --> 12:06.120 That's important to know. 12:06.120 --> 12:07.120 Yeah. 12:07.120 --> 12:08.120 Right. 12:12.120 --> 12:20.120 Okay, let's have a look at this thing actually flying. 12:20.120 --> 12:24.120 Here you can see it land autonomously. 12:24.120 --> 12:28.120 Here you can see it take off by using its propellers. 12:28.120 --> 12:38.120 The propellers, the software is quite similar to quadcopter stabilization. 12:38.120 --> 12:47.120 Yeah. So, once the altitude is reached, the propellers will automatically just fall in when the power is taken off. 12:47.120 --> 12:51.120 And it can fly efficiently. 12:51.120 --> 12:56.120 And after that, we will not use the propellers anymore. 12:56.120 --> 13:00.120 Here you can see it fly for grades. 13:00.120 --> 13:03.120 While it does this, it reels out the line. 13:03.120 --> 13:07.120 And the ground station acts like a rowing machine. 13:07.120 --> 13:11.120 When you put out the line, it generates electricity. 13:11.120 --> 13:17.120 And as the kite can also fly like a normal airplane. 13:17.120 --> 13:21.120 It can fly back against the wind. 13:21.120 --> 13:27.120 And the ground station only has to reel in the other line. 13:33.120 --> 13:39.120 Okay, let's have a look at the open source hardware and software of this thing. 13:42.120 --> 13:49.120 Here you can see the heart of our system. 13:49.120 --> 13:57.120 It's a custom made PCB, including the fly controller. 13:57.120 --> 14:00.120 We have a bigger picture here. 14:00.120 --> 14:05.120 This is just a common ESP32 micro controller. 14:05.120 --> 14:10.120 Very easy to program. 14:10.120 --> 14:15.120 The rest is very simple to quadcopter drones. 14:15.120 --> 14:19.120 It uses the same sensors. 14:19.120 --> 14:22.120 A gyroscope, a very standard gyroscope. 14:22.120 --> 14:35.120 And for the height, it has two centimeters. 14:35.120 --> 14:40.120 And on the right side, you see the power electronics for driving this. 14:40.120 --> 14:45.120 What is new addition is what's in a circle here. 14:45.120 --> 15:00.120 Pressure sensors are also integrated there, so we can more precisely measure the altitude. 15:00.120 --> 15:07.120 That's what's actually in the line angle sensor. 15:07.120 --> 15:15.120 Here you can see the integrated line angle sensor. 15:15.120 --> 15:21.120 We use the gyroscope sensor as the line angle sensor. 15:21.120 --> 15:28.120 Because it has a gyroscope, an accelerometer, and a magnetometer included. 15:28.120 --> 15:35.120 We don't need the magnetometer, but because the kite doesn't care when we're north and south is. 15:35.120 --> 15:42.120 The wind can blow from any direction, so we have a magnetometer for free. 15:42.120 --> 15:47.120 And what we did is attach a small magnet to the line. 15:47.120 --> 15:54.120 And with a magnetometer, we can sense where the line goes. 15:54.120 --> 15:59.120 This way the kite is able to follow the line when landing. 15:59.120 --> 16:01.120 And also, why flying figure aids? 16:01.120 --> 16:04.120 It knows where the line goes. 16:04.120 --> 16:06.120 So it has the angle of the line. 16:06.120 --> 16:10.120 We know the length of the line by counting it on the ground station. 16:10.120 --> 16:19.120 And so we know exactly where the kite is, basically, without relying on GPS. 16:19.120 --> 16:25.120 We just thought essentially why would we actually control this with GPS like basically any other projects. 16:25.120 --> 16:29.120 Because there's all the parameters already defined. 16:29.120 --> 16:31.120 It doesn't move freely through the air. 16:31.120 --> 16:34.120 Because everything happens in relation to the line. 16:34.120 --> 16:37.120 And the line is where we want to generate power. 16:37.120 --> 16:40.120 So why not use that, right? 16:40.120 --> 16:46.120 So what you can see here is our setup for the ground station, which is equally interesting, I think. 16:47.120 --> 16:52.120 Yes, the ground station is based completely on the VEC motor controller. 16:52.120 --> 16:58.120 And open source hardware and software motor controller by Benjamin Fedder from Sweden. 16:58.120 --> 17:03.120 He spent 10 years developing this thing. 17:03.120 --> 17:06.120 It's absolutely wonderful and awesome. 17:06.120 --> 17:10.120 Some people have replaced their motor controllers in electric vehicles, 17:10.120 --> 17:15.120 but this one and got something I 50% performance increase. 17:16.120 --> 17:21.120 And it's programmable in LISP, which is absolutely awesome. 17:21.120 --> 17:28.120 For our purposes, we need to define the line tension based on how fast the line is really out. 17:28.120 --> 17:31.120 And this is just one line in LISP code. 17:31.120 --> 17:34.120 It's just so easy to program this motor controller. 17:34.120 --> 17:43.120 It has also some other advantages like we don't need any additional sensors to count the rotations of the motors. 17:44.120 --> 17:51.120 It just senses all by itself the exact position of the motor. 17:51.120 --> 17:55.120 It's absolutely wonderful thing to have. 17:55.120 --> 18:00.120 And we're really grateful for the open source. 18:00.120 --> 18:04.120 Yeah, thank you. 18:04.120 --> 18:05.120 Yeah. 18:05.120 --> 18:10.120 To the left as well, another custom design PCB was integrated. 18:11.120 --> 18:15.120 Yeah, this is because we need to send. 18:15.120 --> 18:21.120 So the VSE counts the motor revolutions and thus knows the line length. 18:21.120 --> 18:28.120 And then we send the line length using using another SB32 on the ground station. 18:28.120 --> 18:31.120 We send the line length to the kite. 18:31.120 --> 18:33.120 So what can we do with this? 18:33.120 --> 18:39.120 Well, first of all, we can with this setup, we can catch the kite autonomously. 18:39.120 --> 18:42.120 We can take off again. 18:42.120 --> 18:47.120 So as you can see, it's not perfect yet, but. 18:47.120 --> 18:48.120 Okay. 18:48.120 --> 18:53.120 Is it okay? 18:53.120 --> 19:02.120 Yes. 19:02.120 --> 19:07.120 As you can see, it will be just catch by this very simple setup. 19:07.120 --> 19:14.120 And it can take off again, as I said, it's not perfect yet, but we testing it currently. 19:14.120 --> 19:18.120 Yeah. 19:18.120 --> 19:22.120 So we also was a friend from the US. 19:22.120 --> 19:25.120 We also developed a soft kite variant. 19:25.120 --> 19:33.120 So with the soft kite, this will be less autonomous, but as you need probably need to throw it up yourself. 19:33.120 --> 19:37.120 But this positive great potential, for example, like free time use. 19:37.120 --> 19:41.120 And it's going to be much simpler and even more accessible. 19:41.120 --> 19:44.120 And to use this exactly the same drone station as we used. 19:44.120 --> 19:48.120 We told them how to build it and to build it in Montana and the USA. 19:48.120 --> 19:51.120 It's all just ever meeting. 19:51.120 --> 19:55.120 And here you can see him operating this at the Burning Man Festival. 19:55.120 --> 20:01.120 So our ground station is already very proven in hard conditions I would say. 20:01.120 --> 20:10.120 And here you can see basically the features which our kite has for testing as well as operations. 20:10.120 --> 20:19.120 For example, we can edit all the and adjust all the constants of the auto pilot, all the settings. 20:19.120 --> 20:24.120 We can adjust live in flight on our phone or tablet or whatever. 20:24.120 --> 20:28.120 And we can demonstrate this here as well. 20:28.120 --> 20:42.120 And in that feature you can see on the bottom right, which will be more important in the future.