The 'Space Architects' of Mars | The Age of A.I.

Day 217 at the Terra Hab. It's getting pretty bad out there. Winds are gusting to 170 mph. Global temperatures have increased by two degrees in the last decade. Tsunamis are devastating cities. Critical situation! I'm just here trying to survive... Long enough to... -beat my high score... - ...and it seems like... I'm driving towards the finish line... What, did we do? lose the genny? Have you ever had the feeling that we are passing the hours while a disaster looms? Play the violin while Rome burns? We're good in the snow, Trev. Some say we have the technology to fix the planet in a matter of decades, but what if we can't cool our climate or feed a population projected to be 10 billion by 2050? -These topics make me anxious. - Fortunately, there are people who use AI. to address some of our most pressing problems, such as innovating in agriculture and redefining the way we think about food.

And, oh yeah, if things really go wrong, build a new home... on another planet. How can A.I. Help humanity adapt to a new way of life? Anyone in the arena who is not part of the competition, please enter as they prepare to perform the tests. When I first connected it, I connected it to this. -That's when the switch exploded. -Yeah. Then I switched to this cable. That's good. It feels good, it feels good. The 3D Printed Habitat Challenge began as a call from NASA asking anyone in the world, from any background, to contribute architectural concepts that could be created on Mars.

Three. Two. One. And take off. Mars is the next great frontier. NASA has been sending probes since the 1970s and, more recently, even sent a

space

rover to explore the red planet. Confirmed landing. We are safe on Mars. For decades, they have been laying the groundwork for their true mission... By 2033, they plan to send humans to Mars. To go to Mars, it takes about six months to get there, and then you have to wait about a year and a half on the surface for the planets to realign for a very short return trip. The Apollo 11 astronauts were in a small capsule.

Stay tuned for T-1. It had a volume in the lunar module equivalent to a C-block cell on Alcatraz if it were only five feet tall, but you can't expect someone to live in a habitat like that for any period of time. Um, yeah, that's why they'll need a suitable place to live on the red planet. The problem is that it will have to be built by A.I. before they arrive. We have a few minutes, if you could find a couple more cables so we can get power for our electronic devices. - Well ok. - Good?

Arrest. Well. NASA has been holding an open competition for

architects

and engineers to design a habitat that can be built autonomously on Mars. Ready, Chris? Now, the competition has reached the final round, and the winners will receive half a million dollars and the opportunity to join the Red Planet Project. I don't know how many

space

architects

there are in the world. It is not a common job. There's not much work yet, but... my wildest dream is to get us into space faster, but also revolutionize the way things are built on Earth. So for Mars, we know we have to give it the shape of the pressure vessel, right?

And, Jeff, you had investigated, -um, dome... - Right. Bull, cylinder... Cylinder... I am an architect. My specialty is making skyscrapers. I want my company to be the first construction company in space. When you build in space, every little grain of material will have value, so we want something that is as strong as possible with the least amount of material possible. Based on what you just printed, what do you think is actually the easiest build? We can do anything, as long as we have enough time to place the plastic. In the past, you had to use experience and empirical evidence to figure out what to build, which is why A.I.

It is so transformative in the field of space exploration. AI. allows you to study millions of options, compare iterations with each other and find the optimal one. This process is sometimes known as "generative design," a fancy way of saying that machines learn to do what architects do. Here, A.I. The algorithms compare and predict how different designs and shapes perform in a virtual Martian environment. That function could be structural strength, thermal insulation, or simply trying to reduce the amount of material needed to build a habitat. Alright, let's see how this idea works. There you have it, so that's the dome.

But really, what we want to do is... - Ah. - Turn it around. A vertical habitat? How to stack them on top of each other? Yes. Life on Mars is not as simple as Bowie made it seem. Air is a toxic cocktail composed mainly of carbon dioxide. There is basically no gravity and no water. Dust storms can last for months and the temperature ranges from -50 to -200 degrees Fahrenheit. Probably the biggest challenge is the temperature. The habitat has to survive daily changes in temperature, as if you traveled every day from the equator to Antarctica. They're working there, they're eating there, they're sleeping there, they're hanging out there.

What kind of lifestyle do they have? It should keep you protected from that dangerous outside, but it shouldn't make you feel like you're in prison. There are some, we can call them, like, atmospheric gradients here, from loud to quiet, another one might be from dirty to clean, and the third one might be from dark to light. The AI, by doing many, many iterations, discovered a highly optimized result. That concept of this kind of vertical egg shape will actually work much better for several reasons. But Mars isn't the only planet where AI-powered design is finding important new applications.

There's an emerging field called geodesign, which is really about saying how do we take AI, merge it with geography, with civil and urban planning, put them together and you can say, "Well, what happens if we put the wind turbines?" here?" "What erosion will occur if this dock is built in this area?" "How much density should be allowed to build in this area?" "How much water is there? Is it contaminated?" Geodesign has a profound ability to change our lifestyles, especially in cities and supercities. Generative design helped propel David's group to the finals of NASA's cosmic competition in Illinois... - You're the guy? - I'm the guy. ...to face a Penn State team.

Ready? Three, two, one, let's go. As much as possible, everything here should simulate the process on Mars. They have 30 hours to build a one-third scale model of the habitat, which is about two stories tall. It has to be built only with an AI-controlled 3D printer. We can't test a building on Mars. We can afford to send that equipment to Mars, because it's too heavy, so how do we know that what we build is strong enough? And the only way to do that is through AI. AI uses computer vision cameras. to monitor progress and adjust as necessary.

It's the same technology used to print things like auto parts and jet turbines. Well, maybe just do your... -I could do this. Yes. -...back and forth. We definitely want to use artificial intelligence so that the robot or the system really understands what it is doing and becomes aware of the process. So we are in automatic mode? Do you have a stopwatch running just in case? You have the robot, and it's very capable, but it's a little bit blind and it can't think, it'll just do what it's told to do, so we basically need to take Jeffrey and Chris' experience, and give it that. experience the robot so that when it goes up to Mars, it makes the decisions that Jeffrey could have made, it makes the decisions that Chris could have made, it might even know that the best decision is, like, the average of what those two guys think.

I'm going to time this layer. And remember, there is no Home Depot on Mars, so the AI-powered robots will also need to find, create, and process all the raw materials themselves. So we're printing with what we call Mars-relevant materials that we find on Earth, but that can also be found on Mars. People who are really interested in space actually care a lot about Earth too... so the idea of ​​taking advantage of a building has always appealed to me. The material we select to build on Mars is renewable, fully recyclable and biodegradable, and it turns out that material is super strong.

The material is a mixture of basalt, a common rock on Mars, and plastic polymer made from corn. The problem is that you can't find corn on Mars. It has to be farmed... by robots. It will be some time before we can grow enough corn to print buildings on Mars, but we envision farming on Mars. Growing corn on Mars... is that possible? And if so, can we find a way to grow any type of food, anywhere, anytime? It is a climate problem, but we also have to solve the automation part. Our ancestors in America did not bring with them everything they would need for the next thousand years.

They all lived off the land and I firmly believe that this is the solution to colonizing the Moon and Mars. It's extremely expensive to launch anything into space, so instead of bringing all the food and stuff, I think what we should do is recreate that scenario and grow something on site done completely robotically... and that's why it's necessary to have some type of A.I. to be able to make those immediate decisions. This is science fiction, but it's also right at that tipping point of innovation where it's becoming reality. Back on Earth, robotics is already beginning to transform agriculture.

For now, the work is mainly taking place in a small town in the Netherlands, but it's not hard to imagine how far the implications could reach one day. I have a passion for plants. I already had it when I was very little. I grew up on a farm. I wanted to understand "Why does it grow this way and how can I make the plant grow even better?" Leo is a plant guy... a plant guy whose research not only seeks to make agriculture more sustainable, but could revolutionize an entire industry. The area is one compared to 250 in the United States.

That's almost twice the size of Manhattan. Some people call us the Silicon Valley of agriculture. It turns out that this small European country is now the second largest exporter of fresh food in the world, thanks to guys like Leo, who are helping to grow some of the most sophisticated and productive greenhouses in the world. The secret? Vertical farming. Duijvestijn is a tomato producer. The greenhouse here is the size of 50 football fields and is well known for its extremely high output in a very efficient manner. And the results are amazing. These greenhouses produce seven times more tomatoes per acre than a traditional farm.

The efficiency of a greenhouse is determined by many different factors. And they give exactly what the plant needs, no more and no less. But it's not just the roots, it's the environment above the ground, the humidity of the air, the concentration of carbon dioxide. Sensors hidden among plants generate a constant stream of data, including temperature, humidity and soil nutrients. It's about designing the climate and optimizing food. All that information is gathered on your computer to control the weather. It's very important to collect data and see some trends in it... Mmm-hmm. ...so I can do my climate control better.

Do you have to look every day? Every day I check it. It is massive data. You can't just look at an Excel file and see all that data. That simply doesn't work, and that's where artificial intelligence comes in, which uses all that data to control the crop. People from all over the world come to study at Wageningen University, because it is ranked number one in agriculture. The robot is removing a sheet. We generate a model... it is not only measuring the leaf, but also collecting the leaf samples. Despite being at the forefront, Leo and his team are still in the early stages of AI. agriculture.

To help him innovate further, he hired a guy at the intersection of biology, artificial intelligence and robotics. But the magic comes when we start to make sense of the data, looking at how one batch of plants can compare to another batch of plants where they received different treatments, and if we understand the differences between these batches, we can get a better understanding of how to optimize the plant growth. AI. It's what makes the magic, helping them see what they otherwise couldn't. Using ultraviolet and infrared light, Rick and his team are training A.I. to measure photosynthesis in plants.

AI is really helping us bridge the gap between raw data and the desired outcome. Using computer vision, Leo and Rick can measure the light reflected by the leaves. This allows them to see how much energy a plant generates at the levelmolecular. Some of the main applications of computer vision are in sectors such as agriculture and manufacturing. One of the big opportunities is that while we can see visible light, there are many other parts of the electromagnetic spectrum, such as X-rays, infrared and ultraviolet. It is possible to build sensors for them, and this opens up a whole space of possibilities for machines to solve problems that humans cannot.

However, this can only be done when the plants are not too large. To make things even more efficient and impressive, Rick's team is working on an automated robotic arm. With the combination of depth perception and the moving arm, the A.I. It has the ability to recognize leaves, stems and fruits in the dense greenhouse environment. So you can see one leaf after another, or the fruit behind the leaf. Yes. There is a lot of labor in all those greenhouses for harvesting the fruits. We selected bell peppers on purpose, because that crop is actually very complicated. Bell peppers are much more hidden from view, so we generated a virtual greenhouse... and trained an AI. figureFind out which are the leaves, which are the stems, and which are the fruits, and also measure whether that particular pepper is ripe enough to be harvested.

The holy grail is that there will be greenhouses around the world managed by robots and artificial intelligence. If what happens near Amsterdam doesn't stay in Amsterdam, and A.I. Precision agriculture becomes as common as tomatoes or peppers, can you imagine the implications for the human race? My dream for the future is that we can grow plants in a very efficient way, without any harm to the environment and that there will be enough food for people around the world. Leo's dream can help sustain life on this planet. But what about other places? Like Mars? Can we go that far?

Where the weather is more fickle and nuts and peppers are hard to come by? Not to mention moldable synthetic polymers. I have to be your eyes. Alright. Nice and easy. We chose two materials that obtained good results: this biopolymer and this basalt, which is a super resistant compound. The other team is using concrete. One of the main differences between polymers and concrete is that if the concrete has to stop, the concrete starts to harden in the pipes and cannot harden again, whereas with plastic, if we had to stop printing for any reason , you can reheat it and it will return to its original state.

You can see the first row of that layer. That's where that shelf was made. I mean, this is, what, six layers above that? Remember, it's all about adaptation. Because humans won't be there to supervise, the A.I. Robots must be able to observe what is happening, recognize if there is a problem, find a solution to fix it, and fix it. -Up a little bit! -Oh! We want to use machine learning on things so that the system adapts to different circumstances. The same goes for this competition. Since the robots are supposed to work autonomously, the teams are penalized for any intervention they must perform.

The two types of interventions we are dealing with are remote and physical. As long as we're pushing buttons and we're not adjusting the physical system at all, those are remote interventions. That doesn't count as much against us. The robots will be alone on Mars long before humans arrive, so if Jeff and Chris want the A.I. To learn to work autonomously, they need to teach and then step back... Oh no! ...even if they see mistakes coming. Is it falling? A bit. The humans see the problem, but the robot does not. Because we're printing that layer so quickly, it builds up a lot of heat, so it sags in that area.

Well, we have to do something about this now. Let me be your eyes. Just tell me when I'm halfway there... -Okay. -...or three quarters. Three, two, one, let's go. Every time Jeff counts down, Chris slows down the printer to fill that space in the print a little better. Two, one, back. It is a difficult decision. Do nothing and everything can fall apart. Intervene and you'll receive a penalty, each of which reduces your chances of winning. Three, two, one... come back. Remote, remote! Remote intervention. I'm sorry. Jeff's only saving grace is that Penn State is having problems of its own.

It's over there. I think it's okay, right? Yes, I have to go away. Three, two, one, come back. Are you getting better, Jeff? We have to continue doing it at a slow pace. So how many layers are left? We really don't know. It's probably between 55 and 50, something like that. -Fifty five? -Yes, so... Do you think we can do it? In the race against time, we have to reach a certain diameter toward the top so that our skylight can actually sit on top of the mold. If we don't get high enough, it will just fall inward. Three, two, one, let's go.

Penn State is almost finished. Finishing faster doesn't win you the competition, but it counts for something. We need to complete the other cone. That's what's happening now. Well, it looks like Penn State just completed its cone. They are celebrating. Okay, now we only have ten minutes. Meanwhile, Jeff's team still needs to find a roof over his head. Uh, we should continue until the last moment. The top should be narrow enough to support the round skylight that will complete the enclosure. Jeff? -Jeff? -Yeah? Are we printing another layer? - Yes. - No. Chris says no. One more.

We can't do one more. It's 5:54, Jeff. One more. - This is the last layer! -Yes I know. We almost arrive. Tell me when. While Penn State finishes... -Manual mode? -Yes please. ...AI. Space Factory stops the printer to have the last five minutes to install the skylight. Trusting in You. Yes, I'll tell you if it will fail. I can't see it from here, so... I know, I know. If they fail, they will likely lose the competition. We only have one chance to get the skylight, so it's all or nothing. Perfect so far. Well. - Whoa, stop! -No!

Arrest! -Below! - Below. Forward. Alright, put it on top and then press play on the code. Down 18 inches. Nice and easy. Tell me when, tell me when, Jeff. Go on, keep going slowly. Thirty seconds! Keep going. Keep going. Keep going, keep going. Disengage. Could fall. Could fall. Eight! Seven! Six! - - Oh! Fell. The material did not have enough time to dry. The weight of the skylight pulled those... those hot layers, and the skylight... Oh, it just fell. I heard the thud and said, "Ah." For those glorious seconds, he was there, he was there, at the top.

The final score depends on several key structural tests. One of the tests is the smoke test. NASA will take a smoke payload and place it inside the printed habitat. The smoke test is intended to demonstrate that the habitats are airtight... ...but without a sealed skylight... We will be a chimney. With a whopping zero points on that, A.I. Space Factory is comfortably in the... underdog position heading into NASA's other key assessment... ...the crush test. Wow. Mars is relentless. It goes without saying that the structure must be strong and resilient... to survive the competition, yes, but also for life on Mars, because, you know, they have those dust storms that last forever and giant space rocks launching to you. at a million miles an hour.

I don't even know how heavy that excavator is. It's the biggest one I've seen in person. Alright, here we go. Oh! Oh! Nice! Man, wow! Oh Lord! This is crazy. About to crush our habitat. It's happening? Oh! Do you see him get up? The CAT is rising. Look at it? Oh shit! All your weight is on that thing. Oh my god, that's crazy! My God! - Dude! - Dude! Shit! That's a lot of weight! -Oh, those are the... -Oh! - - Yeah! That was crazy! Yeah! That was all amazing! - Hurrah! - That was great! AI.

The Space Factory structure has withstood the crushing force of a 90-ton excavator. Will it be strong enough to support life on a planet that does not have what we might consider a good quality of life? Time will tell. For now, everything depends on the judges. Will this surprising result be enough to convince them? I'm stressed. My... my body can sense that I'm stressed. My back is quite tense. So we're going to award all our money... and... first place goes to the A.I. team. Space factory! It couldn't have gone better. Life on Mars is not possible... yet.

Getting there, much less surviving, is much harder than the Moon, but NASA is exploring it... And now, so are some private companies, like Elon Musk's SpaceX. They imagine a future in which intergalactic travel is not only possible, but normal, like taking a cruise through the Caribbean. Welcome to space. Some even think it's necessary, because if we can't fix the planet... aren't lunar projections like Mars... Leo precision farming... ultimately it's about adaptation, to a changing climate, more people, a plan B in case we can't act together? Or maybe it's just a better tomato. The population is growing and the demand for food is increasing rapidly, so artificial intelligence is the key to improving performance.

We want to be the first people to build outside our planet, and the only way to do that is through AI. If we want to go to Mars, if we want to populate other planets, these kinds of things require these advanced technologies. Space technology is the ultimate test of ideas and ambitions. I would like to take some of the DNA of our work and revolutionize the way things are built on Earth. We should always think of ways in which what we are doing improves the Earth. Hey, I'm going to do the beam with Riedel. -Well. -Alright.

The load test of our material. Specifically, it is a bending test. Are you quite happy with the centering? Yes, it's centered, I think. Twenty-four, 25, so we have to get to 40. We're at 30, we're 75% of the way there. It's deviating slightly down there. Well, we got it, we got it, we got it! - -Alright. At least it happened. Now to see it fall apart. I dare these guys to try to break it. Now you can see it bend. Yes, now you can definitely see the bow. This... is going to explode. Oh man, that's great. He bounced pretty high.

Boom! Is that what it came to? He must have, yes. Ninety-eight. The surprising thing is that our material does not have any steel reinforcement. This is purely Martian material made of biopolymer and basalt fiber. -It's a good one. - Wow.