Latest / The Punch List / Why building on the moon is harder than it sounds?

Transcript

Why building on the moon is harder than it sounds? · The Punch List

• TK (00:00) Does the Moon have earthquakes? Kenny Cheung (00:02) There are absolutely moonquakes. TK (00:03) Will they be in their space suits, or could I be walking around in my pajamas? Kenny Cheung (00:07) We wouldn't have people in a spacesuit. People can be in, effectively, pajamas as well. Jon Wright (00:13) Power, internet — how are we addressing that? Kanav Hasija (00:15) I was reading online that there's ice in the craters. Are you going to use that ice? Kenny Cheung (00:22) The water— Jon Wright (00:30) Welcome to The Punch List podcast, the number one podcast for architecture, engineering, and construction in the world. Exciting episode today — we have a guest, Kenny Cheung, research scientist at NASA Ames, Director of the Coded Structures Laboratory, and principal investigator on ARMADAS, the Automated Reconfigurable Mission Adaptive Digital Assembly Systems Lab. Kenny is working on construction on the Moon, so today we're going to talk about NASA's plans for a permanent Moon base — the construction methodologies, the challenges and opportunities around that, as well as what that means for application on Earth. To set the stage: NASA, the US, plans to have foundational infrastructure in place for a permanent Moon base sometime in the 2030s. There are Artemis missions planned around that, including one earlier this year — that's the stated goal of NASA, and we have great scientists like Kenny working on it. I'll just tell you, Kenny and I know each other from the pool with our kids, and through dozens of conversations I found out Kenny is working on construction technology, and thought it would be a great opportunity to talk about this on The Punch List. So Kenny, welcome. Kenny Cheung (02:00) Thank you for having me here. Jon Wright (02:02) Absolutely. So let's talk about the Moon base. Why the Moon base? What's its intended purpose, and why now? Talk a little bit conceptually about what NASA's goals are. Kenny Cheung (02:19) NASA has an overall goal of doing exploration, and we say exploration for the benefit of all humanity. Part of the exploration we've been after is looking at other planets and being able to send people there. We've understood for a long time that one of the best ways to develop and test out the technology to do that is to use the Moon as a stepping stone. On top of that, there's still a lot about the Moon that we don't know, and there's a lot we think we can do there in terms of scientific instrumentation, for astrophysics or planetary science, that we think would be useful. A lot of what we've done has been these beautiful spacecraft that have to fit into launch vehicles and get deployed. But if we can build up much larger infrastructure — whether in orbit, like the space station, or we have lots of ideas about how we could build even larger infrastructure if we had a base to do that on, like the Moon — that would help us move forward. So for human exploration, it's really about understanding what we need to do and developing the technology to support human activity on another surface. That technology is something that will make its way to places like Mars. Jon Wright (03:48) Yeah. Talk about what foundational infrastructure means on the Moon. And maybe TK, when you think about foundational infrastructure for a construction site in general, what are some of the things you think about, just to set the stage — and then Kenny, talk about how NASA thinks about it with the Moon and some of the unique challenges you face. TK (04:10) Before I start — I was sharing this before we started the podcast too — growing up as a Star Trek fan, I always felt that one day we'd be building structures on other planets. So I really appreciate you being on this podcast, Kenny, and I'm looking forward to hearing how we start with the Moon as a stepping stone. For me, foundational infrastructure — when we're building buildings out here, one of the key things about a building is human comfort, and then fire life safety. Those are the two biggest things that are non-negotiable. Everything else you can negotiate in or out with the client. I have a lot of questions, so I'll start by saying this. The intended purpose of the base is for humans to live there for a certain period of time — is it months, days, or hours? That's one thing I think about. Second: if it's over days, they need to be able to stay in there. Will they be in their space suits, or will the whole thing be contained in a way that — could I be walking around in my pajamas in there? Kenny Cheung (05:25) The long-term intent is definitely sustained — that's something we've been able to do in space. People have been on the International Space Station for more than a year. Jon Wright (05:35) Construction site, I’ll immediately think about temp power, construction trailer, internet — the basics nowadays of getting something going. How are we addressing that? Power generation, comms, et cetera? Kenny Cheung (05:48) That's exactly how we're approaching it. There are these basic categories of needs that have to be filled. We have teams across the agency working on different strategies for installing power, generating power — both solar and nuclear — and putting together the comms systems. In a lot of cases it's about adapting what we currently do on Earth to be able to survive launch and then survive the Moon environment. A lot of the technology development work is about dealing with the fact that it gets unusually cold, and unusually hot, for anything we do on Earth. There's also the fact that there's not really any atmosphere. The dustiness of the environment is pretty much as bad as the worst dusty environment you could have on Earth. So it's a harsh environment for doing anything, but we know construction environments are pretty bad too, with concrete dust and things like that. There's lots of details that need to be worked out. We start off with an idea that's like the trailer, that's like camping, and we want to move it to something that's more robust and provides more capability. TK (07:03) By the way, one side question — does the Moon have earthquakes? Kenny Cheung (07:07) There are absolutely moonquakes. It's actually a big part of the environment. They're not particularly strong, but they last a long time. In fact, some of our industrial partners in the building industry have been helping write the ASCE building code for the Moon, based on what we know of the seismic environment. TK (07:28) Oh, careful — you might need a permit and an EIR on the Moon then. That's the biggest risk Jon and I deal with. Jon Wright (07:36) California cannot extend its jurisdiction to the Moon. Do you need a NEPA permit for the Moon? Question for the next podcast. Kanav Hasija (07:48) I have a question, Kenny. I was reading online that Europe is planning to go to the south side of the Moon, the South Pole, because there's ice in the craters it can use for hydrogen and oxygen. Walk us through the process — how are you going to use that ice, and how does that help with sustained energy, water use, or oxygen use? Kenny Cheung (08:14) The goal of extracting that water is for the oxygen, and it's an energy source as well. The technology to develop that resource — the water — is important both if you want to use it for oxygen for humans to breathe, or as an energy source, which is the most important one. That technology is something we absolutely need to develop in order to do activities on Mars. Part of the idea is that it can sustain Moon industrial activities, but it's also where we develop some of the technology to sustain Mars industrial activities as well. The most basic form of using that resource is electrolysis — you get hydrogen you can use for rockets, and oxygen you can use for rockets as well, or for humans to breathe. We think we can also extract oxygen from the regolith itself. Are we familiar with the term regolith? It's Moon dirt — there's lunar regolith, there's Martian regolith. Jon Wright (09:26) What's the primary component of regolith on the Moon? What's the chemical composition of it? Kenny Cheung (09:31) It's thought of as being a piece of the Earth, because the chemical composition is mostly silica, and it has a lot of metal in it — mostly iron, and potentially useful amounts of a lot of other metals. We have companies working on how to make aluminum parts with lunar regolith as the starting point, finding different ways of extracting the aluminum and forming parts to use in machines. If you just melt it down, you get basalt — approximately the same kind of rock you'd find on Earth, and it actually has nice properties. Mechanically, it's something like the technical glasses we use in the laboratory. Jon Wright (10:12) There seem to be two schools of thought here in terms of construction materials — ship it, or make it, if you will. Your research touches on this, but talk to us about the current thinking in terms of what materials would be used for the structures being built. Will we ship it? Will we make it? Will we do both? Where's your head right now? Kenny Cheung (10:37) Near term, we're pretty much going to ship everything, because we're still figuring out how to make it there. Medium term, the goal is to figure out how to make as much as possible there, using the materials and energy we can derive on-site. Long term, I believe the real technological solution is for all of your equipment to exist in a way where most of it can be reused. In my group, we work on modular structural systems and robotics, where the idea is that the general utility of any one individual component is broad enough that it could apply to lots of different kinds of missions. You can think of it like general-purpose building components that any building could be built out of — with the addition that it's a little bit easier to take apart, without really compromising on the robustness or firmness of the structure when it's assembled. Jon Wright (11:42) Are we going to be using steel, or are we going to be using concrete? What are we using for the structural material? Kenny Cheung (11:49) Short term, steel is on the table, because mass is so expensive in terms of the fuel you need to send it up there. We're talking about things like steel, aluminum, titanium, carbon fiber. There are plastics that are useful in space. To build things early on, we'll probably be using the lunar regolith as a mass-shielding component, and possibly for building foundations — things like berms, the kind of site work you'd expect on Earth. Jon Wright (12:19) Talk about the shielding. Are you talking about grading and berming, what you're referring to with shielding for the regolith? What does that mean? Kenny Cheung (12:27) Or even being able to cover over a structure to provide radiation shielding and micrometeoroid shielding. The radiation shielding is, to a lot of people, the hardest problem. TK (12:39) So it sounds like it'll be some kind of dome-like structure — you ship pieces from Earth to build there, and then put the lunar regolith on top of it. Like in the old days, you'd build a hut out of straw and put dirt on top to make it hard. I mean, not that the Moon has rain, but is that what you're using for radiation protection? Kenny Cheung (13:05) It's not just solar radiation — cosmic radiation is something we need to deal with for humans. It's also nice to have a way of dealing with it for computer equipment. We're going to be sending things like job-site trailers that fit on rockets, and they can't be occupied for very long because of the radiation problem. So we do need to find ways to build the radiation shielding. There's a lot of potential ways to do that. The lunar regolith isn't the only possible way to deal with it, but it's nice because you need a lot of mass to effectively provide the shielding. And so we think the lunar regolith is a good source of that. The most basic way we can think of is what you described — building or deploying something that can support just loading the material on top, and that's one of the ways we're going to do it. In the future, I think it'll be nice — if we're using the lunar regolith for shielding — to turn it into building components that might seem more or less like traditional building components that can be put together, and, let's say, more seismically robust than a pile of dirt on top of a dome-like structure. Kanav Hasija (14:24) So one interesting fact you mentioned, Kenny — and I'm just spitballing numbers here — until you ship it, we're talking about a building, but when you start to make it, it goes down to millions, and then down to less than a million as well. So that's two to three orders of magnitude difference, right? Is that a good assessment from a numbers point of view? Kenny Cheung (14:46) I agree that the physical and economic intuition is that it should cost orders of magnitude less to build a building out of material that's on-site than to ship it all there, for any size building. We're actually still working on the technology to shovel dirt around — that counts as a technology development project right now. We're going to be finding some things in the next few years where the role of robotics is going to be moving dirt around, so even that is sort of early days. You mentioned construction as a problem of materials, equipment, labor, and time. We look at everything as a resource, and try to optimize where all of those resources are in space and time. What are those resources? The building components, and any aspect of the robotics. We want to be able to solve this kind of chicken-and-egg problem — we can't send astronauts to build everything by hand, so what we really want to do is build the systems the astronauts need in order to survive and do the work we want them to do, and be as useful as they can be. We want to be able to do all of that with robotics. The robotics is a really big question — how do you do the robots? NASA has spent a lot of time pioneering the humanoid robot, the humanoid astronaut, and that's still an approach that's interesting. My group specifically works on looking at this question of how you distribute these resources. We look at the robotic resources in terms of all of the actuators — the individual component of a robot being able to move one thing to another place, or install a fastener — the individual sensors you might include on robots, and how that's distributed over space and time. So on one end of things, there are transformers, where all of your robotic resources are everywhere all the time — your building could be a robot, set up so that it can sense everything everywhere and move everything everywhere. There's been a lot of research on robots, and some research on buildings, that work this way. We've taken a step back from that and thought about how the robots themselves can move around the system, and what things really need to be active when. We've arrived at an approach that's like really high-performance Lego. For us, that's broken down into an approach that's really centered around the coordination of the robots. The robots we build in my group are really, really simple — much simpler. Even something that looks like a robot arm, in our systems, is much simpler than robot arms you'd normally think of. Kanav Hasija (17:52) Interesting point — I just want to ground our audience towards like, sequencing that they understand. When we think about the sequencing of building a building on Earth, people think excavation comes first, then foundation, then superstructure. It's a very different ballgame to the sequencing you're dealing with here. Can you explain to our audience what the sequencing looks like? Are you sending payloads first, then robots, then building structures? Walk us through the sequencing of this ten-year, twenty-year plan to get there. Kenny Cheung (18:35) What I was getting to is that the sequencing is important — I like that you're pulling that out. Being able to optimize the sequence in real time is something I think is critical. By breaking the system down in a modular fashion, we put all of the complexity onto the scheduling and planning — the sequencing — and pull a lot of the complexity out of the hardware. By having any individual hardware component be relatively simple, and responsible for a relatively small piece of it, but being able to coordinate everything, you end up with something that's a lot more robust and a lot more scalable than our traditional systems. So what does that mean? It means we think the way we move robotic systems forward isn't necessarily to make robots more complex, but to make sure we can make robots work together — and that coordination of robots is what's super important to us. Kanav Hasija (19:59) It's so funny — we'd love to learn that back on Earth as well, once you get there on the Moon. Coordination is the biggest pain point right now in construction. But you said one thing really well — there are so many other variables and constraints you have to deal with in space and on the Moon. You'd rather simplify the hardware as much as possible and make each robot do one thing really well, and the coordination is the bigger complexity to handle. Kenny Cheung (20:33) Yeah, the coordination is what makes it work, and I think that's true on the job site as well. What you want on a job site isn't more people who can do more things each. You want your experts to be really good at the thing you want them to do, and the problem is coordinating all of your experts. I think that's where robotics and automation in construction on Earth moves forward as well — we're going to have really specialized robots that are really, really good at doing the one thing they do. Kanav Hasija (21:14) Whenever space technology has developed, it's found secondary use cases and impacts on Earth as well. In your quest to build construction on the Moon, have you found any secondary impacts on Earth as well? Kenny Cheung (21:31) We think so. In a relatively near term, there are situations on Earth in the construction industry — particularly for building really high-performance things, whether bridges or towers, certain kinds of buildings, certain kinds of work — where we need to build structurally high-performing work, but the environment is really difficult for humans. We have those situations on Earth, in site development situations. Bridges are really dangerous to work on. Tall buildings are really dangerous to work on. I think we'll find applications in those domains early on, and then with gaining efficiency and performance in these systems, you'll start to see them in more everyday construction applications. TK (22:20) In some ways, to Jon's point — you're breaking construction into the least common denominator of pieces that can fit into many things, so the focus shifts more toward coordination on how to use those pieces effectively. Like in Lego — you've got four or five different kinds of blocks, but you can put them together to make many different things. Jon Wright (22:43) Construction on Earth is a 10,000-year-old industry, and we optimize for material, labor, and equipment — in modern times we're probably optimizing for labor, that's our biggest cost driver. We were going to ask how construction on the Moon is different, but it sounds like optimizing for simplicity, and standardization, may be a guiding principle. How do you think about that? Kenny Cheung (23:11) There are different ways of looking at simplicity. There's an analogy I really like, about the effect of the design problem on standardization, using digital images. We're all looking at screens right now, and anyone watching this podcast is looking at a screen. If you're looking at a 4K screen, for each one of the pixels there's a very finite and discrete number of things that pixel can display. It's an inherently standardized modular system — you have enough of these pixels, they're all essentially the same, and the number of images your screen can display is a number greater than what we think is the number of subatomic particles in the universe. What we can do with what is essentially a simple modular component is really incredible. From an engineering standpoint, that's where we are with digital communication systems, digital computation systems — we know how to engineer systems that are modular and give us enough variability that we can even talk about the artists and designers being satisfied by what's being implemented. That's the way we're thinking about it from an engineering standpoint, for hardware and materials, and in particular structures — the structural systems we're working with don't need to have that many different kinds of components that fit together. They don't need that many different kinds of Legos, and that's why you can build anything out of Lego, even though you don't need nearly the whole set, but you do need a lot of Lego to do it. But that's fine. That can be very efficient. Kanav Hasija (25:16) So to conclude this thread — you're working on simplifying and modularizing construction for space, and putting more pressure on the sequencing of steps and coordination, which I like a lot. I think this can be reused on Earth as well. What's your concluding thoughts on the research you're doing for the Moon base? And more importantly, how can the audience building on Earth read your group's research, get inspiration from it, and stay updated on it? Kenny Cheung (25:45) When we make breakthroughs in the performance of systems, we look for ways to test them, and we look to the broader community to contribute to and validate that. We do technical and academic publications for that purpose, and we do engagements like this. I look forward to being able to talk about this with others, as we make developments on it. I think if there's something we're learning from trying to tackle the problem of doing construction in space, on the lunar surface or the Martian surface, it's that there's a lot of promise in robots, because they can be coordinated. I expect you will have listeners that may develop some of these robots, and that would be great. Many years from now, if somebody comes back and says, “we heard this podcast and we're now the leading robot supplier for doing this particular task on a job site,” that helps make everything faster and more efficient, done better and robust. The other thing is the reversibility of construction systems. If we can figure out how to think about the cost of materials across the building industry in a way where the value of those materials can be retained, reused, or reconfigured — the physics is there, this is possible — you should be able to unbolt your stuff and put it back. The problem is there's just enough non-reversible stuff that happens in building a building — and the labor to undo it is so costly — that we never reuse and recycle stuff, or we rarely do. I don't think it necessarily takes the implementation of all the robotics to achieve that. I think once all the robotics is implemented, it'll be easier to achieve, but there's no reason that has to come first. The more reversible all the construction methods are, the more easily reversible they are. You mentioned it's the second-biggest industry in the world — depending on how you measure, also the second-biggest energy user in the world, and much of that is in the embodied energy of the materials. So it's fairly straightforward to think about not having to reproduce those materials when we're doing renovations from one building to the next, because a lot of the building industry has to move toward the evolution of buildings. There isn't that much space. Jon Wright (28:28) This is great, and we're really grateful for you coming here today. Thank you so much.