המעבדה הסודית של MIT: כך בנו סטודנטים טכנולוגיה עבור JWST ומאדים!

Yep. Good. Well, thanks for inviting me here today. Uh it's I understand it's a celebration of the uh space systems lab which uh is somewhat overdue but uh you know better late than never. The um my passion throughout my career has been doing research or technology research and my other passion is flying experiments in space with students. So this story that I'm going to tell today is really a marriage of the two. It's how to create re technology research laboratories in space that are not just useful to us but actually to a wide variety of organizations worldwide. So that's the story I'm going to go with. This story starts with baby pictures of

course. So if you look at the upper left corner, you can tell that's the old picture because it's in black and white. That's me on the very left sitting on my mom's lap. I'm six months old and this is the first meeting of the Society of Women Engineers in Pittsburgh, Pennsylvania. Yay. Yay, Pittsburgh. Where's the towel? You know, and uh and it happened because my mom was the co-founder of Society of Women Engineers in 1948 at Drexel University and it became national national nationwide in 1950. Anyone a member of SUI? Okay, we need more. And uh I got interviewed by the um by the national office for their 70 70th anniversary newsletter. It was about founding families. And they asked me a question. Did you notice that your mom was different than your friend's mom's? And I'd never thought about that. But I thought about it for a moment and I said, "Really? Yeah." because my friends moms were homemakers, you know, hairdressers, you know, hairdressers, uh, grocery clerks, things like that. My mom worked with Admiral Rickover on the drivetrain for the Nautilus submarine, the first nuclear submarine in the for in the world, I guess. The uh she uh would build museums. She did um renovations of old houses. She uh would keep me home from school to watch the Gemini and the Apollo launches. Um the school would call up and she'd read them the Riot Act that I'm getting a better education at home. She uh bought us both motorcycles so we can go riding together. One day she said, "Get in my truck." She had the truck and uh we're driving along and I said, "Where are we going?" She said, "We're going to the airport. It's time you learn to fly." So that night my dad comes home. I'm all excited and I say, "I'm going to go learn to fly." And he said, "Yeah, this is a good time for you to learn how to pay for it." So that sort of suppressed things for a little while, but not forever. So that was good. She said,

"You have to go to MIT and you have to go to Aerrowastro because of their role in Apollo." She would write about space travel, but she was just born too early for that. The um so I went to MIT, Aerrowastro, got my bachelor, my master, my bachelor's, masters, and doctorate. And then when I finished up, I had three job offers. One was uh faculty at Cornell, one was at JPL, and one was research staff at MIT. And I thought I want to work on space hardware. So nothing against Cornell, but I said that not going to Cornell. I figured if I went to JPL, I'd be a small fish in a big pond, and that really wasn't what I wanted to do, but going but Ed Cwley in Arrow had uh recently won a shuttle flight experiment, and I said I could be a big fish in a small pond on that one. So I accepted that research position. the um that really sort of I felt like fired my my career out of a cannon. And uh it's nothing like have in your first few years of doing your your uh you know after your your education of of working with the the astronauts and the and the shuttle program and the research and the industrial partners and all of that to really be exasperating and stressful but very rewarding to do. So um so that really launched my career. The uh Ed Cwley had uh given me two pieces of advice. One was whenever opportunity knocks say yes even if it's very intimidating. I said no to a couple things and I stopped doing that. I started saying yes and that was very rewarding. The other thing he suggested was you should reinvent yourself several times through your career. So I did. So, I spent a lot of my time up here at MIT and and I still do. Uh I still have students, the um graduate students, but I joined the Air Force Scientific Advisory Board. I was there for four and a half years. I got I got interviewed for chief scientist of the Air Force. Didn't get that job. That's all right. Keep trying. I got the uh but I got the NASA chief technologist position and then I got the aerospace position and now I'm at JPL where it doesn't snow. So, that's great. And the one thing that I find very key uh along this process since I've left MIT is that I keep my grad students. I currently have two PhDs and two master's students and that's solving technical problems with the best workforce in the world really keeps me centered and able to deal with the bureaucracy that these other career paths have take have to endure. So okay. Okay. So at the time I became

faculty member, Ed Cwley was department head. He had come up with this conceive, design, implement, operate context for a um undergraduate education and I really embraced that because I thought that's what I want to do. I want to get hardware in space but you got to start somewhere. You got to start with formulation which is really conceive and you got to test it and build it out on the ground and then you got to figure out how to get it launched and then operate it on orbit. So my research was really guided by this. on the conceived side. Did a lot of integrated multifysics modeling, a lot of architectural trade space analysis and actually started a unique uh three semester design build class where we could try to build things and often we learned what not to do but we would also learn what to do and then we would bring that into the graduate program and and up onto orbit. on my design work, I had really mostly studied uh dynamics and controls, but I didn't really want to dive deep into the control side. What I wanted to do was see how controls could take advantage of the unique dynamic aspects of the problem to which we're applying it. So, I'll show you some of that as we go, but here's a listing of the various research threads in uh in the space systems lab. um you know from vibration suppression for space telescopes. All of this really focuses on space telescopes, adaptive optics, machine vision, information flight, things like that. Um and then I want to mature them through test beds both on the ground and on orbit. And um because there's nothing like an experiment to tell you what you forgot in the theory. So that's that's a uh that's a real awakening to go you know the model said it should work and why isn't it working? So, and the other thing is for the facilities that go up to station or shuttle, we wanted to really re really uh operate them like they were a uh a research laboratory on the ground that there's really no distinction between it. And I'll show you some of that later, but I think a great analogy is the wind tunnel. A wind tunnel is where you take a formative technology. You know, we don't quite know what the final form is going to be and we test it in an authentic authentic environment. In that case of wind on orbit, it's microgravity and we want to test it under nominal and more importantly off nominal conditions and without doing harm to the test article, the operator or the or the facility. Um, for our payloads, the underlying requirement was do no harm when you go to space. So um the reason why it's more importantly is if it works under nominal conditions that's good that's necessary but it's not sufficient. You need to know how far away from the cliff you actually are. So by moving it to the point of the technology breaking not the hardware breaking but the technology breaking that tells you how far you are from the cliff and how to extend that distance and that's why you want to do these kind of tests. Okay. So, um, as the title said, you

know, this is 30 years of the, uh, of the space systems lab. So, I'm going to go by decades, but I'm actually going to start in the decade before the the space systems lab started. This is what I call the shuttle era for our lab. It was the space engineering research center. It was 1988 to 95. And um, we flew a number well the first shuttle experiment that we flew was mode. We flew that in 1991. stands for mid deck zero gravity dynamics experiment. We designed this to be very modular in the sense that um there were pieces that were your generic lab equipment or like your data analyzer, your your um storage media, your power amplifiers, your signal conditions conditioners and they were all located in one box and we had that in a mid- deck locker. You can just put it there. But on the front there was an interface that you could attach different types of payloads to it, different kinds of test articles that you could test. In this case, we were doing fluid slush on the first flight. We're doing nonlinear trust structure dynamics and microgravity on that flight. And um and it went well, but much to my surprise, the flight programs kept coming. So in 1994, uh we had the opportunity to refly the hardware but but add additional things to test on it. So, Professor Newman had just joined the department around that time and uh we said, "We have some hardware and it's all qualified. If we just build the things you want to build, we can put it on and we can fly." And these are the dynamic load sensors which are um six degree of freedom sort of load cells on mounted on foot loops and hand holds so that we could get a measure of the most unknown disturbance source on station and that is the cruise motion and how much uh how much Mike think would disturb that lovely microgravity environment up there. So we flew that um and then and then before we got that off the ground we NASA approached us and said hey can you add active controls to the problem? We said sure. So that became the mid deck active control experiment. We tested uh that's shown on the figure on the right here. The uh that went up on the shuttle in 1995 and we did robust um and adaptive control on that but we also did verification through integrated modeling. And I put a star here because this is something I want to I want to talk to talk about in more detail because little did we know that in 1995 when we flew this mission, it would be the pathfinder for how we val how we verified that the James Web Space Telescope would work prior to launching it. So the um the idea may I looked up one of our old journal articles and saw what the objective of MACE was and it was to develop tools for spacecraft that cannot be tested on the ground in a sufficiently realistic onorbit environment and then to have and those tools should allow us to have confidence in the eventual onorbit performance. So that's what the verification process is for flight programs. We had built test articles uh for the shuttle that um changed their dynamics between 1G and zero G. Some we couldn't even test in 1G because it would have broken. Uh and then we would the process we used here is this integrated modeling uh bullet here. We built a 1G model that had multiple physics. In this case, structural statics, dynamics, sensor and actuator dynamics, control algorithms, and time delays and and violin modes and suspension cables and gravity sag and all these things that represented the 1G test environment. And then we did tests and these were control. We designed controllers and implement that. And nothing like a control system going unstable. that give you a sign that something's wrong and you didn't cap you forgot to include something in your theory. And then once we refined that process, we took that we we took the the difference between the data and the 1G model predictions and created an uncertainty model. We propagated that into zero and then applied it to a zero model and design controllers which we implemented on orbit. We also had the opportunity that I think was non-existent at the time for shuttle to bring down data, redesign controllers, send it back up in the morning mail and implement it again over the course of two weeks. I think we did that like five times during the mission. So what does that have to do with James Webb? Well, here's James Webb in the big A chamber down at uh down at JSC Johnson Space Center going through its sort of all up testing. So give you a little bit of indication of what's what. There's the mirror. That's probably the obvious part. Uh here are all these suspension cables because it's 1G. You got to lift it off the floor. Um there's a there's a one hertz isolator that goes between the bus of James Webb, the spacecraft bus, and the telescope. One hertz uh isolators and 1g sag about 10 cm. That would have ripped the isolator apart. So what they had to do is mechanically short it. This is also a nested thermal chamber. So what you're seeing here is a liquid helium wall. And outside of that is a liquid nitrogen wall that keeps the liquid helium liquid. And um but the temperature environment was not really what it was going to see up at Earth's sun L2. So that was a little different as well. So what what we did what James Webb program did is they developed a stop model. It stands for structures thermal optical performance. It's a different it's some of those uh physics are different than what we did in mace but the integration of it was the same thing we did in mace and uh that model was built by my graduate student who worked on mace he built it for the for NASA Gddard and that's the process we used the same thing 1G model the 1G data get an uncertainty model apply that to the zerog uh do apply that to the zerog model and then we would uh predict whether James web was still in back. It was we launched. The other thing is I sort of bookended this because as Dave mentioned, I was on the the product integrity team for James Webb and my role was to advise on dynamics modeling and and uh testing how you would test the dynamics of James Web and so we sort of had both ends of the program there and and uh that was great to see the results of uh mace be applied. So, what I'm going to show here is a movie from the shuttle, and hopefully the sound will be good. Not too loud, not too soft. >> Okay, so welcome aboard. We've got about

a six minute video to show you here. We'll bring you back on the flight deck for a little bit at the end of it. Here we are on the mid deck and I'll show you just a little bit of the assembly of the uh MVP or multi-body platform that is the main attraction of the MACE or MIDI deck active control experiment that we ran uh on Endeavor here for the last 16 days almost. And here I'm uh assembling two of the struts and an electronic uh data connector to one of three nodes. Uh the node in the middle has three ray gyro assemblies connected to it as you can see there. These are also >> over to the ESM or experiment support module which is housed in one of the lockers and also includes the disc drive where the data is stored and the data gets across on the white umbilical that you can see there. Also in the frame right now is the kiss box that we've been using all for getting data to and from the ground. It's that blue box uh on the locker that you see. The MVP consists of four struts, the three nodes, three ray gyros that are mounted orthogonally. And then at either end there is a two two axis gimbal assembly. And the end behind my head there is the secondary or disturbance end. And the other end on the left of the screen is the primary end or the end that we're trying to control. The whole idea here is to put a disturbance in to the end that I have my hand on there uh that actually shakes the whole structure and will in fact uh disturb the performance of the primary end. And the idea is through active support or active control of the assembly on the left to actually allow that or enable that gimbal to point very accurately at at a single point in space. And this would simulate a spacecraft that had say a scanning device at one end and a pointing device requiring great accuracy at the other end. Here we're starting a a run that has a disturbance going in at the far end. And we'll be trying to control the left side here. And the control has just kicked in. And we'll give you a couple of close-up views and and give you an idea of what this really looks like. And the vast majority of the runs that we ran uh were a lot like this. We'll show you a few that didn't work out like this, too. This is the disturbance end, the secondary gimbal and you can see that shaking the whole assembly and in fact at the primary end which is not being driven just being disturbed by uh the disturbance end you can see it shaking here and then the control will kick in right there and you can see that it settles down uh very very nicely. And while we were doing this, we were uplinking uh new controllers, down linking the data, and the folks at MIT were looking at that and they were writing better controllers, and the performance of these controllers got better and better as the flight went on. Here I'm mounting a laser pin to the primary end so that we can demonstrate for you just how uh remarkable this active control is of this device. Here the disturbance has gone in and you can see the laser spot against the fleet station and the control goes in right there and very accurately points the primary in. Now not every single controller worked this way. Some of them uh performed and like I said the vast majority of them uh performed very well and got better as the flight went along. But there were some like in this case you can see the spot against the police station is being driven at a high frequency and is actually being driven off the target. And uh and this is not the kind of control that that you'd like to see. But it does demonstrate that if you're not careful, you can actually make the problem worse by designing a poor controller. And you can make it bad enough that it actually completely overcontrols the structure. As you see here, >> he has to land the shuttle two days later. So we're a little nervous here. We had several different configurations of the uh multibody platform. Here's configuration one, and it's actually the uh the configuration that we ran uh the majority of our runs in. And this is another one of those controllers that did not perform as as well as as you'd like. On the other hand, we were learning an awful lot from these controllers that went into these divergent states. And then the final configuration that we had was actually the most fun. It was configuration two, the original configuration that we showed you plus uh what we called flex appendages which simulated say uh solar arrays on a space station or a spacecraft of some type. And you can see those uh at the orbiter's z-axis along the orbiter Z-axis in the mid deck. And here here's another one of the controllers that actually drove the whole assembly into an over control or out of control situation. And uh the amount of travel that this particular vehicle would have taken was actually limited by the tethers that you see there where it uh would have gone flying on into the uh into the overhead end of the lock space forward of the assembly.

>> So I want to remind you of two things. This was all done by graduate students in an era where universities and graduate students really didn't get the chance to fly things. So I think you know this was a millennium ago. Remember this is in the 90s. And uh the um the second thing I want to remind you of is the wind tunnel analogy. Don't harm the test article. Don't harm the operator. I'm looking at the astronaut. And don't harm the facility. Yet you have to push it to failure. So that was a tricky sort of design constraint. So um and it worked. So I'm going to fast forward to

the next decade. The uh this started in with the founding of the SSL, Space Systems Lab, and the flight opportunities just kept coming. The um we uh got the opportunity to fly two different missions up on the Mir space station. This was part of the uh Russian US collaboration and preparation for ISS. And uh so the mode hardware here, we built on that. We flew what was called uh with Professor Newman the EDLS for extended duration uh dynamic enhanced dynamic load sensors. Thank you. Um and uh and the nice thing there is I remember we were doing it on the shuttle where we might only had two weeks or something and someone said you should see how the Russians fly especially how they adapt over time. So uh but you need a longer time constant to see that. Um, we also flew a version of mode up to do the the Mazu Misdi experiment looking for Javier. Did I get that right? And that was to get an acceleration survey of the um of the Mir space station to understand what its modes were and things like that. So that wasn't with our lab, but the idea of the modular design allowed us to sort of get these things flown fast and and less expensively. Uh we also got the opportunity to refly Mace and this went up on the uh on the space station uh for expeditions one and two. So this is Bill Shepard who's a uh a graduate of course two here at MIT mechanical engineering and I think the space in space station consisted of two modules at that time maybe. Yeah. So it was pretty crowded in there. Um and I'll show you some about that. It's also the year when we started this multi-semester capstone class where we wanted to give juniors and seniors the experience of the full life cycle. Let them come up with a concept and then build it and then go operate it someplace. And the first one was called spheres, the first class project. And right after we finished the class, um we uh we got the opportunity to build it for space station. And uh this was one of the flight units and it has a little expansion port here that would allow us to in the future add other payloads. The thing is we were packed onto the shuttle. I think it was SCS 116 which was supposed to be the mission that was right after the last flight of Colombia. So that caused the whole fleet to be uh to be grounded and um and part of that we had to figure another way to get up to station. So I'm going to talk about here a little bit about the the capstone class and our mace reflight. So this is actually why it sort of looks distorted here is this is out of an IMAX film and this is Susan Helms who was on expedition 2 and uh nice little clip. We we had modified spheres to add more flexible appendages here so we could give it more nonlinear uh behavior. We also modified the uh software to allow neural nets and we teamed with with University of Michigan on this and with AFL and this was the first crew interactive experiment on space station. So thank you all you taxpayers out there. We really appreciate that. Um >> oh that's Mace. Sorry. This is Mace reflown. Yeah I got spheres on the brain a lot. That's coming up. Um >> the mace experiment is how Susan entertained herself on the weekend. Yeah, I really liked working with this experiment. It was like playing with a big toy. It had moving parts and a brain. When you launch a satellite into space, the control system is the designer's best guess of how it should control its own sensors and appendages. This experiment was trying to understand how you could design a control system to teach itself to work better. So, as I said, this is also the uh decade where we started design build capstone class. It started as a three semester class and then it shrunk to a two semester. But I already mentioned uh mentioned spheres here. That was our first project and and David Newman and I co-taught the class. I hadn't taught a class before, so thank you for that. Um I I list to some of these some of the follow-on impacts. It did go up to station was on station for 14 years. Uh it was the first IVA free flyer and number of other nations have copied us now. The well we copied ourselves with the as with the Astrobes that are up on station now the Japanese the Japanese Simon and there's one from issa >> and I think the Russians have been talking about it. So so we went first and I guess it's popular so that's good. Um it also some of the work we did impacted uh the Cassini mission. Some software got loaded from spheres loaded onto Cassini. Makes you nervous when your graduate students are writing software for a mission around Saturn. Uh our fluid slashwork impacted uh Delta IV heavy cruise dynamics during um during its GTO out to out to go GTO transfer and there's a big STEM impact here as well. Uh two years later we did a multi- aperture uh fo interpherometer test bed on an air bearing with control to sort of point the stage control from ACS up to optical control and one of our panel members Chung here is professor Sununjo Chung here was our TA and and that was a great success. We then went to one of these crazy ideas called propellantless formation flight and the way we did it was using electromagnetic um forces and torques between vehicles to uh to be able to do that. that went to ISS and three companies have been bu built around this idea. Fortunately, none of them do I have any ownership interest in. We then built a modular uh spacecraft with docking ports so we could reconfigure them. That went to space station. Our docking ports went to JPLED. Uh the uh we built a modular rover for extreme terrain access and the vision system here went up to ISS. We also built uh two different uh orbital transfer vehicles. Uh then we partnered with uh EPS or course 12 Earth atmospheric and planetary sciences to do a collaborative class. This was with professor Sarah Seager to do to build something she called exoplanets which would be the first sort of 3U cubat that had subarchse second pointing accuracy. Uh when we finished the class and the graduate students sort of revved the design, uh JPL then hired our graduate students, renamed it to a stereo, and it's been it's stereo and it's been flying for a while now. Um we also built an instrument called the regalith X-ray imaging spectrometer which went on the Osiris Rex mission to Bennu and since it's bolted in place, is that right Dr. Mastersonson? It's now heading on its way to asteroid Apous. And uh and then our last project that I was involved in was um a a spinning aperture telescope, a rectangular telescope that you spin to fill in the image plane. And the students chose to name it after our president of MIT at the time. So the first day of class, don't worry, this clip is less than 30 seconds. It's okay. the uh the first day of class um actually before the first day of class Earl Murman told me you know so you want to do this crazy class of trying to you know do this this life cycle experience and I said yes and he said well you know this is the smallest uh junior class we've had in a long time you need to get a critical mass of students involved in the class so I had an idea I think I think Dave and I had the idea is we're going to bribe them so we prom sent out an email to the juniors we promised if you join this class you're going fly on the zero gravity plane. So, we got enough students and uh so on the first day of class, you probably remember this movie, but um uh the uh we we put we forwarded to this clip and I and you what you're going to see here, we want three we told them we want three of these. We want them in station and no laser beams, please. So, we started to work. They all booed when I hit the stop button on the VHS tape layer, but um so we did this and uh we got up on the KC135.

When I first came to MIT, I had no clue that I would be doing anything as cool as this. Never would have thought that I would be doing this ever. >> It's been definitely the culminating experience of MIT for me. This opportunity came up and it was just like, wow, I want to do this. This is what I want to do. It's a completely different real attitude to learning about aerospace engineering. It's giving you a taste of the real world, I guess, before uh before you get a chance to get out there. >> The spheres project is really sort of two things rolled into one. One, it's part of a research program that the space systems lab is running. Second, it's also an academic I've lost my hair, but Dave, I don't know what's going on. >> Undergrads systems engineering. >> We're good. We've been teaching them the design process and it manifests itself into an actual flight hardware and the specific hardware for spheres is actually formation flying satellites. >> We basically interviewed industry and we said what is it you want to see in your new hirees and they said we would like to see hirees that have the full life cycle experience in developing an aerospace product. We're trying to make our conceive, design, implement, and operate the context of an engineering education. >> So, we came up with uh this course, which is really a three semester course and started basically with a clean sheet of paper where we gave them a problem, something to build, gave them some requirements, and they took it from there, built the thing, and are now operating that facility in NASA's ZeroG aircraft. The students were presented with a basic concept and allowed to run with it in their own pursuits. >> The concept is that instead of building large expensive satellites, we can make them cheaper by perhaps building smaller ones and coordinating their activities. One of the key uses of these formation flight technologies is for future space telescopes. >> You could have autonomous satellites spaced far apart and the application there would be interpherometry. interferometry basically instead of having a large mirror takes two smaller mirrors and moves them further apart. Now the angular acuity on the sky is no longer determined by these diameters but is determined by this separation. >> Compared to the great Hubble telescope that's very large. You can now think of smaller satellites but getting the same optical resolution. We can take smaller, less expensive mirrors, instead of the 2 and 1/2 meter diameter of Hubble, move them out a kilometer. You can imagine the kind of fine detail in the sky. We could start to see on the KC135. The real goal is to see if we can predict how our algorithms will work in an environment other than that in which we designed them. So, we can't do that in the lab. We can't test in zero G, but we can design controllers, go up on the KC135, see how they behave in zerog, and see how our predictive capability is. >> They did pretty much what they were supposed to do. You lift it up and let go, and it stabilizes itself. So, we we demonstrated that, which was one of the goals of the of the flights. >> I think that we got some good data on both what the spear is doing and on what the plane is doing. Uh because the plane does some pretty extreme stuff and that what the plane does also affects how our experiment works. I was telling somebody the other day that if we're in a group walking somewhere where I have images of the right stuff floating in the back of my mind and everybody I tell about the project their jaws just drop and they say wow only at MIT and that's the way I feel. And that is not the Brad Pitt but uh thank you.

So the second decade um things really kept going actually it was busy decade and it was out almost too busy but uh we finally got spheres up through a combinate we launched on almost every launch vehicle there is in the world's fleet. Uh we launched on uh the Japanese HTV, the European ATV, the Russian Soyos. Yeah, on Soyos they don't tend to carry payloads and uh progress vehicles and then shuttle once it became operational again and we we actually because of that expansion port uh we made up sent up a lot of different payloads. Some were from the capstone class and others were not. Uh so you know we did proximity operations just because we're operating in an aluminum can we cannot go more than 3 meters apart but there's a lot of interesting things you can do. We gave it eyes stereo vision. I'll show you a little bit about that. The uh we put up uh slash tanks at uh this is a project that was supported by Kennedy Space Cent's launch services providers because they were having some issues with the Delta IV heavy um fuel stabilization. Uh we were also put up docking ports because we wanted to work on inspace assembly. Uh and this was more of our um sort of our bat belt or whatever you want to call it. So you could put robotic arms and docking ports and multiple payloads on at once. And here is our electromagnetic formation flight vehicle. You might remember um it had rings on it. Well, this is actually called rings. It can not only control their relative motion via electromagnetic forces. Uh but you could also do wireless power transfer between them like you do with your phone when you put it on a recharging puck. So I'm going to show you some things about vertigo and rings. That's what the asterisk means. and our UDPs which are universal docking ports and things like that. We also um one day Dr. Sotero came into my office and said he'd been talking to astronaut Greg Chamatov who had operated spheres up on station and said we have a plan. I'm like uh oh and they said you know you the way you do safety was way we do safety with spheres is that we make sure that software is not a safety control. So because that's how the technology is instantiated and we don't want to have to go through NASA safety every time we change software. So Greg Chamat said so anyone can program these like yeah where you going with this and uh so he said let's run a competition a robotics competition on orbit which would be the first one run off the planet. So we called it zero robotics because first robotics already had a bit of a brand and we wanted to get ahead of them in the yellow pages if anyone remembers yellow pages. But, uh, I'll show you a little bit of that as well. But one of the things I really wanted to do was support robotic assembly because if you're going to do robotic assembly, especially of telescopes, you're going to have a lot of pieces you got to move around. If you do that outside and you lose some pieces, you might be asked to testify to Congress. If you lose them inside, they bounce off the wall and the astronaut hits the pause button and we can test again. So, it gives you that time to iterate and take a little risk. So, here's a concept for doing inspace assembly, and it uses our electromagnetic formation flight or EMF, which stands for oomph. We thought that was kind of cute. So, you you get a rocket, you write your name on it, you stack your mirror segments like uh dishes in a uh in a kitchen cabinet, you get one EMF vehicle up there, and you have another one mounted at the at the bottom here. So it's it's magnetically pushing and torquing against the one at the top. You use your UDPs or your docking ports to come in and grab a segment and you bring that out with hope hopefully not scraping the mirror. So magnetic forces, you know, opposites attract. If this is a di a dipole magnet, opposites attract. um the same type of of D of poles repel, but if you move them perpendicular, they'll shear and you can and you can get all six relative degrees of freedom. And so it's propellantless propulsion, but it's all staying within close proximity, which is what EMF needs. And it's all really it's all the relative degrees of freedom that matter. So for you telescope people, you really don't want a flat mirror. you'd like it to be a little curved, but uh we didn't convey that to the animator here. The last thing you do is you grab the uh secondary mirror and uh jettison that truss and hope it doesn't come back to hit you and then you position that secondary mirror up in front of the telescope and you magnetically trap it there and then you get rid of the uh if you remember that secondary support tower you saw for for James Webb, the defraction effects of that are not there. Um, of course there are other challenges here, but uh it's it's possible. So I'm going to show you how for that

EMF, which was a class project, the pathway it took to space station. So we got the class together. This is what they built. Looks kind of mean. That's an open jug of liquid nitrogen. Be careful. Uh, it froze all the water out of the atmosphere of our lab. We got 10 turns of high temperature superconductor in each each of those two coils. And the one in the back is um is bolted down and it's got 100 amps. That's 10,000 amp turns. And the one here we're putting 10 100 amps through it. It's a handful of del batteries. It's amazing stuff. And um here we're just doing open loop. You can see the uh the the nitrogen boiloff coming down. That's not a good trade with propellant, but we actually developed a cryogenic heat pipe that could do that with no consumables. The um so it can push away as well. Those are 100 pound vehicles and uh and they uh it's pretty agile here. I don't know if you saw in the last frame the uh student was wearing his safety goggles up on his forehead. He's more interested in fashion than safety, but Oh, there we go. Yeah. Uh we had to give it an umbrella because it started raining on our electronics. Probably a problem we wouldn't have in in space. But um yeah, and when we changed the current, we just recharge the the del batteries and then we then we pump up the co the coils again. Shear is the interesting one. That's where you put the magnetic dipoles perpendicular. So in this case, the dipole of this vehicle is across the table. We had to commutate it. So it's the currents in that coil, that edge on coil, then that edge on coil, just like a DC motor would work. And the reason it has to rotate is because it's all internal. Remember this channel your undergraduate physics class. It's all internal forces and torqus. But here we've induced angular momentum into the array and angular momentum has to be conserved. So it goes into the spin of that vehicle. So now we turn on a reaction wheel to to absorb that angle momentum so that the vehicle spin didn't have to do it. That's a math error. It's nice to show the physical instantiation of a math error to the class. They didn't design the the reaction wheel strong to have enough torque to prevent it from from rotating. We also always run the camera, always video because you never know when your last test's going to be because the smoke came out or whatever. So video hides a lot of faults. So then we the class graduated and then we brought it into the uh graduate program, learned a lot, redesigned the vehicles and you'll notice they look a lot slimmer now. We added active control to them. We use the spheres measurement system which uses ultrasonics synchronized by infrared. And now this one's under active control. That one's bolted down 10 um 10,000 amp turns in that coil. And so now he's going to push it forward. It puts current into this face on coil to repel. Comes back, but then it has to reverse that current in order to stop its its drift away. You can see how anxious the students are. It's probably the hundth take of this, but uh especially this guy's grip right there. But uh I can imagine the silence in the room. The uh then shear, this is the one where you make the dipoles perpendicular. We do have a reaction wheel that's better designed. could have used a little tuning, but uh it's bringing it back to the uh to the objective position, the center of the table. And um he's very happy. He's he's sort of questioning things. Uh Uh and then they're as strong as um as reaction wheels. These are these are pretty powerful. Um so I'll cut that. So now we got to a point where we sort of had all the pieces coming together for this telescope um robotic assembly lab up on station. So we had EMF and you saw the photo of it on it on station. We also needed things here's just a photo of the team but um they deserve all the credit. The uh the uh you might we might need to be able to dock and undock and if you dock with something you don't know its properties well. You might have to do a system identification to learn them. So we uh we did some of that up in station, you know, for 14 years. This was really a a lab that we used almost weekly up on space station. And um you might need to uh you know, it could be a crowded workspace. You might need to be able to do collision avoidance. So here we command the spheres to go to the vertices of an equilateral triangle and then go down to bis sectors and they automatically detect they're about to collide. So, they have to replan how they're going to go around. And you know when you do that dance on a sidewalk where you're going left, right, trying to pass the person coming the other direction. This is a 3D version of that dance. And then you probably want to have eyes on the construction site. So, we gave it um we gave it eyes. And here we're actually doing a two vehicle inspection of a tumbling object here. One of the interesting things is our propellant tanks on these spheres have liquid CO2 as our as our propellant. That's a whole another story. So this is a mutating spin. One of my grad students saw how that mutation effect um the pole everyone remembers the pole ho right yeah thank you. So um we could see that the pole which for oiler dynamics should stay constant was oscillating and that means it's not a

rigid body. It's flexible. It's the fluid. He was able to estimate the amount of tank amount of fluid left in the gas tank just by watching its motion. So now we had all the pieces together and then we moved to another decade. We actually didn't get that test bed up there bas uh the spheres after 14 years were really running out of gas literally but the thrusters were things were wearing out. So um the third decade here of SSL still had a lot of action going on. We got the opportunity to build Rexus regul x-ray imaging spectrometer on the Osiris Rex mission. I'll show you more about that. You could ask what do we have to do with Perseverance rover landing? Uh, a lot because the JPL entry descent landing team for Perseverance was comprised a by a lot of SSL grads and I'm going to show you that. And then um we also did star shade trajectory work. This this is the next great observatory in in astrophysics at NASA. And the idea is to be able to get spectra of Earthlike planets around nearby stars. But to do that, you got to block the light from that star because it's 10 billion times brighter. So an idea to do that is to put a star shade in front of that in front of that star and it has to be about 10,000 kilometers ahead of the telescope. It's kind of what happens when you when the moon gets in front of the sun, you do a solar eclipse. This is just sort of a human-made eclipse. But you can imagine at 10,000 kilometers in advance, you want to look at other stars, you got to move this thing around the sky to block block other stars. And NASA had looked at how much delta V and fuel they needed to do that. They assumed the gravitational dynamics in Earth's sun L2, which is a threebody problem, are trivial and we can ignore them. So they they did an estimate but my but two of my grad students came in modeled the whole threebody dynamics and developed the optimal trajectories and I was able to they were able to save 25% of the delta V and 40% of the fuel in the mission and um and they really wowed JPL on that. Maybe embarrassed them a little. That's good. Um so I'm going to show you about zero robotics here but I'm going to start with Rexus and Perseverance. Fenway Park may be the home of Boston's most beloved team, but another is about to steal some headlines. This group consists of students from MIT and Harvard who are getting the chance to build an instrument that will help NASA's Osiris Rex spacecraft explore the asteroid Bonu. This group of student scientists are building Rexus, the Regalith X-ray imaging spectrometer. Its purpose is to collect and image X-rays coming from the asteroid to determine its elemental makeup and distribution of elements across the surface. Each student has a unique task on the project. I get to practice being a thermal engineer and honing my thermal engineering skills. >> My role is to work with the solar X-ray monitor as well as some of the avionics side >> on a daily basis. I write test plans and assembly procedures. I work on the Rexus avionic system and what that is is the electronics that take the X-ray signals from the from the asteroid and make sure that they get back to the ground for the scientists to process. As a systems engineer, I'm responsible for making sure that all the different technical disciplines come together. Thermal, structures, avionics, software, they all come together to make one working instrument that meets our science requirements. and making sure everything stays in check is Becky Masterson, the Rexus instrument manager and the group's faculty supervisor. >> Rexus is an interesting instrument on Osiris Rex in that we're a student collaboration experiment. The goal here is to educate students and and the Rexus project has been a fantastic way, you know, to do that. Both undergrads and graduates go through this project and get a real a real live view into what it is to build space hardware. And building Rexus is an intense process with strict contamination controls and elaborate software and hardware testing. And there's one thing every college student can relate to. Note takingaking. >> Uh on Rexus, the reason that we have such careful procedures and note-taking is that in the future if during an environmental test or other integration activity we notice an anomaly, um we can try to trace it back to a root cause. And if you have detailed notes, it's easier to go back and figure out what may have happened during assembly. >> So, what's the biggest challenge in building Rex's? The answer may surprise you. >> Having students involved in Rexus has been a a wonderful experience. Um, it's been an adventure. One of the difficulties of it is that students are always coming and going. So, you get a student in, they're here for a two years masters, um, and they learn a lot and they're doing great and then they're gone. So working on Rexus is a bit like riding the tea in Boston. Some students going outbound, some coming inbound, but all are getting the same collective learning experience. So the biggest thing I've learned while working on Rexus is really that that instruments like this and spacecraft like Osiris Rex are very very complex. There's a lot of dependencies, a lot of requirements, a lot of relationships, a lot of design decisions that all impact one another. Rexus provides a hands-on experience for students to build flight hardware for a NASA mission and that's incredibly rare that as a student you can build something that's going to fly in interplanetary space. So this is like a leadin to a future career in NASA, a future career in rocket science. That's exciting for me. Uh I get a kick out of thinking that, you know, these these pieces that I'm touching and and this assembly that I'm building is is actually what's going to fly in space. So, I think that's pretty cool. >> And between Fenway and Boston's other attractions, students must feel completely spoiled with their social life, right? >> Yeah. Well, uh, I don't go out much. I don't know. What is there to say? Usually, I have to work on Rexes. So, while it's true these students may strike out trying to find that perfect balance between school work, the social life, and working on Rexes, as scientists, they are hitting a home run for the Osiris Rex team. So, I'm running a little long here. So, I'm uh this is a photo of a bunch of uh grads, graduate students that uh graduated from the space systems lab and they're the core of the interest the entry descent and landing team for perseverance at JPL. Do I have 10 more minutes? Okay, then we'll show you the nice thing is they put this is all SSL students. one is a social member of the SSL, but um they were uh uh they put this video

together about a month before the landing. That's confidence. >> Oops. Hold on. >> Nothing can be taken for granted when you get to Mars. There's a lot of things we just don't know. >> Space always has a way of throwing us curveballs and surprising us. >> I mean, until we get the data that says we're on the ground safely, I'm going to be worried that we're not going to make it. Entry, Entry, descent, and landing is often referred to as the seven minutes of terror because it takes about 7 minutes to get from the top of the atmosphere of Mars to the ground safely. >> The spacecraft has to do all of this by itself. >> There are many things that have to go right to get Perseverance onto the ground safely. >> There's a lot counting on this. This is the first leg of our sample return relay race. There's a lot of work on the line. Starting about 10 minutes before atmosphere entry, we get rid of really the spacecraft part of of the rover that's been supporting us. >> We come screaming in to the Martian atmosphere at 12 to 13,000 mph. And the heat shield is what dissipates all that initial energy through friction. >> The vehicle will continue actually flying itself through the atmosphere. It's sort of like a transforming vehicle that went from spacecraft and now it's kind of like an aircraft actively guiding itself. When we're going slow enough, we deploy a parachute. >> The biggest supersonic parachute we've ever sent to another planet. It's critical for slowing down the vehicle. >> Perseverance's entry, descent, and landing borrows heavily from that of Curiosity. >> But fundamentally, Perseverance is a different rover. She's bigger. She has different instruments. >> We've added a lot of smarts on the inside to make it more capable so that it can deal with the landing site that we've given it. The science team identified Jezro crater as basically an ancient lake bed and one of the most promising places to look for evidence of ancient microbial life and to collect samples for future return to Earth. Uh the problem is it's a much more hazardous place to land. You look at Jezro, all you see is danger. How do we go to a site that we never thought was safe enough to go to before? So the heat shield, which has protected us all the way through entry, is no longer necessary. We need to get that off so that we can actually see the ground. And we can see the ground in a couple different ways. Perseverance will be the first mission to use terrain relative navigation. So while it's descending on the parachute, it will actually be taking images of the surface of Mars and determining where to go based on what it sees. This is finally like landing with your eyes open. Having this new technology really allows Perseverance to land in much more challenging terrain than Curiosity or any previous Mars mission could. amongst the rocks and the craters and the cliffs. These things are hazardous to the rover, but these are the things that are interesting to the scientists. >> Once Perseverance has figured out where she is, jettison the back shell and parachute and light up our rockets. Those rockets help us steer to a safe landing spot that's nearby. >> That descent stage takes us all the way down to about 20 m off the ground. >> That's when we start the sky crane maneuver. And once the rover has hit the ground, the descent stage will cut loose from the rover and fly away to a safe distance. Surviving that 7 minutes is really just the beginning for Perseverance. Its job, right, being the first leg of sampler turn to go look for those signs of past life on Mars. All that can't start until we get Perseverance safely to the ground. And then that's when the real mission begins. So I was a little bit hurt that I taught them how to build telescopes and they go land rovers on Mars. But I like that. So I changed my mind said I hope I taught them how to think. So that was good. And actually three of my uh women graduate students are were leaders still are uh leaders on three major telescopes that were very high-tech. So at least at least it's stuck some places. the uh so

I'm gonna go back to zero robotics here and uh as we were as this was happening this was Dr. Sotero's brainchild. As this was happening, Nat Gio got wind of what we were doing and they wanted to make a movie. This is the trailer. And I apologize if I cry in this, but >> I want to go to space. >> You want to go to space? >> Yeah. >> Oh, okay. This is the first time I'm hearing you say that. >> Your mission is to program satellites inside the International Space Station. That is the space race right now. >> And it's going to take every single one of you to make sure that our code is one of the best in the nation. >> Robotics teams have been working in a worldwide challenge using coding skills to see who can do it best. >> The spheres experiment that's about to take place on the space station. You can dream whatever you want. And if you can work together very well, you'll fly straight to the moon. >> We're going to win this space race. >> Okay. >> My goal is to become a marine biologist one day. The water is just like zero gravity. Close as you can get. If you even reach space, you'll be like, "Oh, we're the champions and stuff." >> The early years are the crucial years for them to get exposure to different opportunities and to find out where their passion is. >> This is a robotics competition. Which of your TN's code will it be? >> Seeing something that's in space that we're programming on Earth, that's like kind of crazy for me. >> Let this be the start of a journey, a lifelong journey of learning. >> Whenever people say the sky is the limit, I just think that they're wrong. There's much more ahead of them. T-minus 10 seconds 9. >> Good luck to all of the contestants and let the games begin. >> So, I want to close with uh thanking so many people that helped. Uh, in white are all the grad students that I could remember, but I'm sure Becky and Alvar will find some that I missed. The M means masters, the D means doctorate, 50 doctorates, 94 masters, and I'm not retiring. We're still going. Uh, but I really want to thank the people on the right for Alvar who created spheres and zero robotics. Becky who led the team to fly the furthest student build experiment to AU sorry Marilyn who made sure that we grew up from kids to being responsible adults Paul Bower who taught things that you can't learn in a classroom and one day he actually said we can tuck tate we can duct tape down the circuit breaker on a high power amplifier and it worked. Dave Robertson who loaned us so many tools and I don't think we returned them. Todd Billings who kept who kept us safe despite all our exuberance. Ray Sedwick who brought a real spread to the lab both intellectually and communitywise. Javier Deluis whose company impedance matched us to the professional world of space flight. Ed Cwley who taught me a lot and uh taught me about uh how to wear steel toe boots when chasing funding. That's a good lesson. Dan Hastings who opened a lot of opportunities, taught me how to speak truth to power. Jeff Hoffman for being Jeff. I remember he invited me several times to a private meeting to the commander of Apollo 15, Dave Scott. That's a rare event. You don't say no to that. And my wife, Mindy Morren, for put for putting up with me. And uh thank you all. And I I successfully avoided Q&A. Right. That's the goal. That's the goal. >> Sure. >> Sure. Any questions? Any questions? Questions? Questions? Anyone Anyone have funding? >> Thank you, Al Tadros from Redwire. Um, and class of 88. So, uh, as we're looking to go to the moon, uh, and create a permanent presence there, uh, what experiments, what do students get to look forward to doing on the surface of the moon that we might not have been able to do on station or shuttle or other platforms? Have you thought about that? And maybe your new role? Uh well my role at JPL you know since we got there first if everyone remembers the ranger and surveyor missions had a little trouble doing it but finally successful you know I think there's some interesting challenges uh well there's many challenges but I think yours was more about the research question uh but I'm going to mention one challenge I I believe uh we need Dave Gus Scott here right now when those landers when the line when the lunar modules landed the uh the blast ejected the dust at orbital escape velocity. So, we got to think about what we're going to do to the orbit around the moon as we come in with even stronger rockets and how much of that debris is going to get into space and how do we deal with that or prevent it. Um, terrain relative navigation is a big thing for landing. They did it on Perseverance. Um, uh, but the, uh, lighting conditions on the moon are very challenging, but to get So, I'm doing the DC pivot. I'm answering the question I want to answer. So back to your question, I think a lot of it's going to be in the area of uh life sciences. So that's a great question for David Newman who has left the room just in time. But uh oh, she's back there. Okay. Um you know, there's a the uh the the quiet of the the backside of the moon, you know, it's it's about the only radio silent spot in the vicinity of Earth now. and uh doing uh astrophysics radio or whatever out there. It would be uh great research against that sort of science. But the robotics of building that kind of telescope in a lunar crater is very interesting. The uh how to uh how to um extract ISRU, you know, the the water from the frozen the the sunless craters in the South Pole. I saw a great I remember when I came back from NASA headquarters once met with oi and he showed me this work by one of your students and it wasn't about how do you extract ISRU it was about when is ISRU useful and buys itself into the architecture I have found no one who has answer who has answered that question other than oi and his student and I think that's going to be key into figuring this out um the Uh, I I'm out of time. Sorry. Hope that helps. Thanks,