Student Projects Modeling the First Landing on the Moon | MathWorks Research Summit - MATLAB & Simulink
Video Player is loading.
Current Time 0:00
Duration 13:48
Loaded: 1.19%
Stream Type LIVE
Remaining Time 13:48
 
1x
  • Chapters
  • descriptions off, selected
  • en (Main), selected
    Video length is 13:48

    Student Projects Modeling the First Landing on the Moon | MathWorks Research Summit

    From the series: MathWorks Research Summit

    In a two-semester exercise lab, computational science and engineering students modeled the first landing on the moon. They read original NASA documentation to model the lunar module designed during the legendary Apollo program. As it is not easy to land on the moon, the original solution to support landing was to use a computer. The challenge was to understand the hardware of the lunar module and to control it during descent. The students first tried their own solution to land. However, after finding the pitfalls and problems of their solution, they tried the original NASA approach and learned how to solve such a difficult problem. As a result, this presentation gives insights on how historical engineering problems can be used to educate students using today’s engineering software. The talk introduces the Apollo spacecraft navigation and guidance system. It shows the principles of the Apollo Guidance Computer and how it supports the moon flights. Then it provides an overview on solving the Apollo problem with Simulink® and demonstrates how NASA engineers solved the problem. A landing simulation programmed by the students closes the presentation.

    Published: 17 Mar 2023

    So I'm very appreciative that you will allow me to talk a little bit about that. The original idea of this comes from a few years ago I saw the film, In The Shadow Of The Moon. And there's an interview with Buzz Aldrin in it and he told from this 1202 program alum in the first landing on the moon.

    And I'm teaching embedded system design, and so I make a little bit research what happened there. And this is a nice thing that's exactly what we're doing in research. Embedded, real time systems analyzes, and the origin of this is the first moon landing.

    So NASA and GPL investigated later what happens in the computer, and the trans theoretical model of that is now the foundation of all what we're doing. There was a publication and that shows us. So the idea was now, maybe, could we reproduce the landing on the moon by students with modern computer technology?

    So this is one of the famous pictures, the lander together with Earth. And you can also reproduce this by using Simscape. So this is the lanena from NASA. It's a CAD model downloaded from the NASA web pages. Earth comes from Google, and the moon comes from Google.

    And you can put all together and putting it in the physics, and now you can land this lander by yourself on the moon. So in the past times, they have a big bunch of simulators. These are some photographs from the simulation and wireman given by NASA to train astronauts how to flow to the moon because this has never been done before.

    So you see you need a whole bunch of equipment to do that. So they use a big computer facility to do that. And there are additional photographs, and it's amazing.

    And I remember I was five years old. Not exactly. I was four years old becoming five years old, and I remember this moon landing. So it's one of my first remembers that I was with my parents at the television and see it.

    And then you came up, if you grow up, you think about computers. And then, the first pocket calculators came. And there was also a small game landing on the moon. And so I think the idea to become an engineer was this moon landing because I saw that as a five-year-old child.

    And as you can see, they simulated on the computer like we have later in the 80s, our computer games. So they can train the astronauts on computer models. And looking what happens if they are . This is credit to a nice television series, The Moon Machines that you can find this footage from the simulating of how they do it landing on the moon.

    So now coming back to teach engineering. So normally, this is one of the slides Silvio has also in a different style that you have students are bored in the lectures. And if we want to do engineering, I think the most impressive thing, if you want to become an engineer, then you want to build something.

    So the idea must be build something. So I want to be an engineer to build moon machines, so just for short. And so if you give the opportunity to build something, let's build it. How could we do it on the moon landing?

    So first we try out Simscape. This was a big, cool experience because you have all automatically these reentering like in a computer game, and you can see what you are doing. So let's see. This was a model. It was originally designed by three students working together in a project team.

    We have in our curriculum the opportunity to do projects. This is one semester. They have half of the load of the semester to spend in the project. And they built that model. And at the end, you get this.

    And as you can see, you can fly the lander. You can also see what happens. And it goes down very nice. And the problem is you have just one engine. And you have balance, the deceleration to stop, and also the landing. But if you stop, you fall faster.

    So as you see, this was manually done. And if you make a mistake, it goes up again. So maybe some guys know six months ago the movie First Man where Neil Armstrong is 15 and coming down to Earth because the atmosphere holds in.

    But if you are not trained and you try that, after 20 choices you think, how horrible. This was hard. This was really hard to do that. So modeling is one thing. On the other side, it's a little bit difficult. We want to have a real model.

    And we try to have variable masses in Simulink. And there was, at that time, no support for it in Simscape. So we change back to Simulink to have a variable mass because the lander is burning fuel the ship becomes larger. So we want to model this, as well.

    So we go to Simulink, and there is a nice thing. And I'm very proud to meet him at this event. The one thing is you model that. And Richard, 20 years ago, makes for the 30th anniversary of the moon landing Simulink model.

    And he wrote on the web page or on the paper, he wrote about that, that he needs one hour to model. And then, everything was fine. So OK, students do that. And then, after that, they model everything. It's fine. They need not one hour. They are not an experienced engineer.

    After two weeks, they came. Model is ready. We can do that. We can land. And we can that-- making that. Oh, but how to land on the moon? Professor, tell me how to land on the moon.

    OK. I have this experience from Simscape. And I know 20 tries, 30 tries, hours on experimenting, and not a chance to land on the moon. No idea how to land on the moon. Let's look what Richard has done.

    So he figured it out. And he said that he works from 1963 until 1966 for three years there. You figured, oh, three years on moon landing. So OK, how to do it in one semester? So oh, it becomes hard. So let's read what he has done.

    And additionally, I made a little internet research, and I found how to solve complex technical problems. If you know nothing, look to the internet. You can do modeling. Then, looking to the internet.

    This is a question. This is one of the documents of NASA. You found how to do it. This is more in the conceptual stage. As I was a child, all the books about engineering and science were full of such pictures.

    So there exists a document. This is the title page of this document. And I think it's 200 pages where they describe how to land on the moon. So the students were not funny about that to read 200 pages about what is going on.

    And as you see, the first sentence is we are 10 people working a couple of years to figure out how to do it. So this was really frustrating for the students. And now it comes the best thing for what an engineer can do and scientists can do.

    Don't be frustrated. Try the challenge, read, and look what happens. So they need then three weeks and then got it. But they read that. So in this document, you find a completely description at which point of the fly you have to do what you have to do.

    So you see all these angles, and the highs and so on. And how the land must be in its range. But OK, if you see this page, what to do? You need a little bit more information. You look to other documents, and you're going on.

    Then, you find out, OK, I have to know how the engine is working that I can control it completely in the complete range of the force. So they figured out a lot. And at the end, it comes up to this graph.

    And they model now the last phase of the landing doing that in Simscape. No. Doing it in Simulink. This is the final model. You can go inside and find a lot of stuff about the spacecraft, the Earth, the moon, and so on. And at the end, you can run that.

    So compiling is a little bit faster because about the limited time here I'll make this a little bit faster. But now you can see how it lands automatically. And you must turn this diagram. What is called-- that ax is the surface of the moon. And you must turn it 90 degrees. It's going up now.

    And the secret to land on the moon is this diagram where you see the difference between the angle speed and the angle difference, which was developed by Grant. And as the students figured out how that works, they can successfully land.

    So I want to give a lot of credit to Grant, who solved our problem. No, we solved this problem again. But he does pioneering, and they land also for the moon. So let's land. Took a little bit. It's a little bit faster now. In original time scale, it needs more time. But you can, if you like.

    And if you want to do this on a normal PC, you can do it now in real time. Not a big bunch of computers, not a real big facility with its own power plant. Not a bunch of many engineers. You can do it with this nice mathematical software.

    A lot of inspiring energy doing the work as a student who does not know how and to read what they have done. So I think now we're coming up, down. And let's see. There is only a little bit, so I can skip now.

    These are the results if you plot on different diagrams in Simulink. So at the old, we have three groups of students. We have had three groups of students. Two ones who made the Simscape model for manual landing. One has made the Simulink model.

    And later on this winter term, two students working on the landing. And now, the idea for the next students is to bring this landing from Simulink, maybe, back in Simscape and maybe write a function, which allows us to have variable masses. So I'm very excited about that.

    And I hope I can build a lab for embedded systems design engineers, which is based on the story of the moon landing. So thank you very much for your attention.

    [APPLAUSE]

    Related Products