The Changelog: Software Development, Open Source - Building an open source excavation robot for NASA (Interview)

Episode Date: December 11, 2019

Ronald Marrero is a software developer working on NASA's Artemis program, which aims at landing the first woman and next man on the Moon by 2024. How Ron got here is a fascinating story, starting at UCF and winding its way through the Florida Space Institute, working with NASA's Swamp Works team, and building an open source excavation robot. On this episode Ron tells us how it all went down and shares what he learned along the way.

Bandwidth for ChangeLog is provided by Fastly. Learn more at Fastly.com. We move fast and fix things here at ChangeLog because of Rollbar. Check them out at Rollbar.com. And we're hosted on Linode cloud servers. Head to Linode.com slash ChangeLog. This episode is brought to you by DigitalOcean. DigitalOcean is the simplest cloud platform for developers and teams

with products like Droplets, Sp droplets spaces kubernetes load balancers block storage and pre-built one-click apps you can deploy manage and scale cloud applications faster and more efficiently on digital ocean whether you're running one virtual machine or 10 000 digital ocean makes managing your infrastructure way too easy head to do.co changelog again do.co slash changelog. Again, do.co slash changelog. Welcome back, everyone. This is The Changelog, a podcast featuring the hackers, the leaders, and the innovators of software development. I'm Jared Santo, Managing Editor here at Changelog. Ronald Marrero is a software developer working on NASA's Artemis program,

which aims at landing the first woman and the next man on the moon by 2024. How Ron got here is a fascinating story, starting at UCF and winding its way through the Florida Space Institute, working with NASA's Swampworks team, and building an open-source excavation robot. On this episode, Ron tells us how it all went down and shares what he learned along the way. Let's do it. All right, well, we're joined by Ronald, don't call me McDonald, Marrero. Ron, thanks for coming on The Change Log. Thanks for having me, guys. You got my nickname right.

That was awesome. I figured it's a nickname that you don't want. We'll just perpetuate it to all the internet so that everybody calls you that from now on. And consign you to that fate. Anyways, don't call him Ron. I'm like, I don't call him Ron. Ron Marrero's here. And we're here to talk about a pretty cool project.

Something near and dear to Adam's heart. NASA related. That's right. It is the Easy Razor. Yes. It's called R-A the EZ Razor. Yes. It's called R-A-S-S-O-R, which is an educational robotics software platform built by students at UCF, yourself included, Ron. And in conjunction with the Florida Space Institute and NASA,

Swampworks engineers at the Kennedy Space Center. Well, that's a mouthful. Tell us about what's NASA Swampworks, first of all. Yeah. So Lockheed Martin has this famous lab, this prototyping lab called Skunkworks. Okay. And so NASA decided to model that same energy and create a team of other engineers to do primarily rapid development. And so they created Swampworks.

And if you know about Florida, we have gators just about everywhere. Swamps, yes. Yes, and Kennedy Space Center has a lot of swamps. They've even reported panthers on center. And so Swampworks just, yeah. It just seems fitting that they would have chosen that name for their group.

Yeah, it's like Skunkworks, but Swampwork. Makes sense. I like it. It's actually really on point point too especially considering how much the everglades like those are just a tremendous size of the florida area and i've actually swam in a lake in florida at night what like a crazy person yeah i was young that don't any better yeah that's really dumb don't ever do that i lived that is classic as you can tell or demand ron tell us you're smarter than adam well yeah you know i've only swam in lakes during the day i like to see what's going on they have a really neat logo

it's it's like this blocky gator you know right around the word nasa nice so yeah they've done a good job with that so tell us how you got involved. You were a student at the university of central Florida, I assume a recent grad, give us a little bit of your backstory into getting involved with swamp works and this very cool robotics platform. Yeah. So I know you're of my bachelor's of computer science at UCF.

And so part of the requirement to complete your bachelor's is to go through senior design, which is this really neat group project that you have to complete in order to basically receive your degree. And it's a whole, it's several weeks long where different sponsors will come out and pitch projects. This for me was fall 2018. Tons of different sponsors there were Red Lobster was a sponsor, Lockheed was a sponsor. And so tons of cool projects for students to get involved in. And so the nature of them, the more I started listening to them, you know, they were interesting, a lot of websites, a lot of machine learning as a hot topic. That's definitely one that a lot of students were interested in. And then I heard NASAa there was a gentleman by the name of mike conroy on behalf of the florida

space institute where he works who brought us this project and i remember clearly one of the slides said if you work on this project you will be a nasa intern you will get access to kennedy space center and so that drew me in And then he explained that it was a robotics project and would involve things like a simulation and communication autonomy. And then that drew me in. And so I raised my hand at the end of that presentation. And I asked probably the best question anybody had asked, which is, is there travel reimbursement for taking this internship? I don't know why that came to my head, but that's what I asked. And the answer is no, sadly, but I was still drawn enough by the project to take it. And you go through an entire kind of vetting process with

the professor of the class. He wants to make sure that students who want to be part of a certain project are going to perform well in the project that they're applying for and have the skills and have the resources to do it. And so I applied and about less than a month later was accepted on with nine other guys to start this completely open source robotics project. Pretty cool. What did that feel like when you got accepted on? Were you like, dang, I got to go buy a plane ticket or whatever sort of logistics? Or were you just excited? Did you celebrate with friends and family? What kind of a, what level of honor was this kind of big deal, a small deal? Oh, this was a big deal for me. I live right now in a smaller

town, Kissimmee, South Orlando. Everybody knows Orlando, not so much Kissimmee. And so the biggest things I had ever seen were Disney and times I'd gone to Puerto Rico. And when I found out about NASA, I honestly couldn't believe it. I hit F5 on the webpage that showed I was part of the group a few times just to make sure it was real. And it was. And I am still in shock and all that I got to be part of that project. I immediately called my parents. And then within five minutes, I had a flood of texts because they were so excited and

spread the news. And yeah, it was just a really, really wonderful experience and definitely nothing that I had ever expected to be doing. Kind of modern day equivalent of pinch me. I'm dreaming is like F5. I'm dreaming, you know, there's like hard, hard refresh that sucker, make sure there's no cash. Is this possibly real? Yeah, it's funny. So everybody knows NASA. I mean, the name itself is very iconic in space and we just celebrated our 50th anniversary. And so I was very excited

when I found out about it. And then the longer I was on center and the more I researched after I got the offer, the more excited I became in space. So I think in a lot of ways, I'm way more excited now than I was even back then. You mentioned in what you learned here that you had to write a strong cover letter. What was involved, I guess, in writing that? What was involved in the research to make sure you got it right? Yeah. So I definitely wanted to make sure that I was competent to perform, that I would stick to my commitments, that I would show the professor that if I got a project as important as working with NASA, that I would be able to deliver. And thankfully, I had been working at a hospital group here in Orlando, working software development for them. And they had given me the opportunity to be the scrum lord over one of their development teams. And so I really drove home the leadership

part of my experience, and then just displayed a willingness to learn. I think I've always done a good job at searching. And if I get stuck, just work in the problem. I had no robotics experience. I had no simulation experience. All I had was the desire not just to work for NASA, but to work on something I never had before. And with robotics being such a big part of today's world and an even bigger part going forward, I really wanted to do something that would get me out of my comfort zone. So how long was the duration of this internship, the project? This project ran all the way from October 2018 is when we really started getting into it until May of 2019.

And so in between there, we got to go to NASA several times. We made it a point to go every Friday. And so that's where the Swampworks engineers came in. So they were the subject matter experts for us. I should probably explain what we were asked to do for this project. So NASA, the Swampworks team, they've designed the RAZR. It stands for Regolith Advanced Surface Systems Operations Robot. And it's a digging robot. It's a digging robot to go to other planets, primarily the moon and Mars and perform what are known as ISRU operations, in situ resource operations. And all it means is digging up, being able to dig up the topsoil

and another planet that has elements like hydrogen, right? And being able to dump that into a processing facility where you can convert that hydrogen into fuel. The way they explained it to me is when you go on a long trip cross country, you don't take all the fuel you need with you, you can't. And so it's going to be even harder to do that going to Mars, you either have a big tank or you find ways to mine and make your own fuel. And so the Razor project is just one way to do that. They have a second version of the Razor. They've been promoting it a lot. So we were asked to do an open source version of their robot. Why open source? So their code was proprietary and their project was proprietary. Because of the nature of the missions that they were doing, they could, of course, talk about their design and talk about what they were doing.

But they couldn't publish any code and they couldn't give people who were interested in the project anything to work on. If a school wanted to, for example, make their own version of the Razor, they're not going to have the funds to be able to build the industrial parts that they're using for their robot. They're also going to have to start from scratch. And so when Mike Conroy with the Florida Space Institute brought us this project, he wanted to make a robot that would educate others on the work that Swampwork is doing and allow them to participate in that learning. And the only way that we envisioned doing that was through open source, through being able to have open discussions and being able to share our code and also to work from the ground up in a black box to make sure that we develop something that was completely not developed by NASA that we could

freely share and that other people could collaborate on. Did they give you any sort of bootstraps or leg up to begin the project? Anything like research papers or anything to say, well, here's where we've been and maybe resume from there and make it open source? Or how did it translate from some proprietary version of it to what you all created? Yeah, so they had given us public links to their scientific papers on the Razor, how they had designed it, the kind of challenges that they were working to solve with it, and then just how they engineered it from a high level. And so we went through those to understand, okay, this robot has two digging arms. How did they get that to work?

Where is the central processing unit on this robot so that we can make sure we get that right? What are the kind of operations that they expect it to do so that we can emulate those functions as well? And so we got those documents. We also, we just looked them up on YouTube and were able to thankfully find visual footage of what this robot looked like before we

were able to meet them in person. And we went from there. Interesting. Did you have to learn anything about, let's say like environmental challenges, you know, like extreme weather changes, high, low temperatures, anything with physics or gravity or any, anything that was like sort of outside of typical computer science learning you might have done yeah so this robot it handles it handles movement and obstacle detection obviously through cameras and sensors and so one discussion that we had that we kind of got stuck on was the martian landscape and so there are like dust storms on mars and if you have a camera that needs to be able to see to know where it's going, how do you tackle that challenge of basically cleaning the lens as it's moving along?

And so we, like I mentioned, we got stuck on that and then kind of moved away from it because it's more of a stretch goal outside of the scope of what we were trying to do. But along those lines, we were talking with them about their real challenges. Mars was just one example. Like you mentioned gravity, Mars and the moon have very different gravity fields. And so you need to account for that. The Earth, 9.81 meters per second squared, that's not the same number on either of those planets on the moon or Mars. And so it became a ton of rabbit holes for discussion, but also really eye-opening for us in terms of these guys didn't just make a robot. They're not just working on a robot.

They're working on real space challenges that when they're solved, it's going to make for an amazing continuation of the project. And I'm sure they've solved most of them already. And they have even more challenges ahead to work on. So you successfully navigated this program. Are you now working with, you said when you mentioned NASA, you said we, are you working there now? Have you graduated or is it still ongoing? Yeah. So I graduated in May and now I work on the Artemis mission. And so NASA's, the big marketing push now, and it's a very real mission, is to go back to the moon by 2024.

And so I work with a contractor at Kennedy Space Center to do the software engineering for the ground systems related to that mission. I do still keep in contact with the Swampworks engineers, but I don't actively work with them. I am still keep in contact with the Swampworks engineers, but I don't actively work with them. I am still a maintainer. I actually am an admin over the Easy Razor projects. And now there's a whole new group of students for their senior design that are trying to push along even further than we did. Can you bring us up to speed on the Artemis mission,

what that is and what's involved? Yeah. So we have a lot of missions right now that are in low Earth orbit. I'm sure you've heard of SpaceX. They just had a launch earlier this week and Boeing as well. And so we haven't gone back to the moon in decades. And the moon is important for us to go to Mars, right? And so Mars is very far away. It's a place we've never gone, certainly not with humans. We've sent rovers over there, but we need to get it right. We need to get it right the first time. And so the Artemis mission,

the tagline is we're going back to the moon by 2024. And that's important so that we can test out ideas. We as NASA, I mean, can test out ideas, can prove out ideas in terms of building a habitat on the moon so that when we get to Mars, we can get it right. And so that all happens through a vehicle or rocket capable of taking the biggest payload we've ever taken. It's called the SLS or the Space Launch System. And that's going to be the rocket that we use for Artemis 1 to take us in orbit around the moon and return. And so the Artemis missions are, at least initially, Artemis 1 is to orbit the moon and then come back and eventually, along the way,

by 2024, to send humans back onto the surface of the moon. What I find it hard to believe is you have just blown my mind because a lot of people may perk their ears up and say, well, we, you know, we haven't been back to the moon in quite a while. And there's a lot of skepticism around the fact that we've gone to the moon or whether we've even been in space and, you know, to hear, you know, a mission to say, we will go back and land with a man and woman on the moon in 2024 is

like shut up a lot of skeptics, I suppose, and prove that we actually have gone. Cause I mean, there's a lot of skepticism around what happened in the seventies going to the moon and whatnot. So I'm excited about that. And, uh, I love space thing. It's really interesting. And, uh, I find myself in that weird section, intersection of skepticism around the moon, or at least going there. Yeah. And, you know, traditionally, projects like these aren't always the fastest projects. And so we're in 2019. And to say we're going in the next five years, it's going to take a lot of work and a lot of coordination. And it's a very daring statement to be able to make. And, you know, from what I've seen, we're totally getting there.

I think it's going to take a lot of coordination across the country. There's not just Kennedy Space Center. There are tons of centers across America. And we're really making a push to get it done. And I, for one, would love to see it. I've never seen, based on my age, I've never seen anybody go to the moon. So to get to experience that, yeah. Me either. Me either. I wasn't alive then. We've only seen footage. It was live back in those days, but Jared and I were both pre-moon or post-moon. Yeah. My parents don't

even have memories of it because they were so young. So this will be, I think, a really defining moment for our generation to be able to go back and to watch that happen. I agree. How cool is that to be a part of that mission? I mean, that's so, I can't even comprehend it. That's how cool it is. What an awesome path to go on to, like starting with this Easy Razor internship and then transitioning that into this contract position on the Artemis program. Yeah, it takes a lot of hands.

It takes a lot of moving parts. And there are a lot of real challenges that have to be solved. You know, we're working on the ground systems, which relates to, you know, making sure the rocket can get in the air correctly at Kennedy Space Center. And that's just one part of the mission. You know, what we do on the moon is another part of the mission. The kind of experiments that we run and challenges that we try to solve on the moon are entirely different.

And so through projects like the Easy Razor, where we can give code to university students, high school students, even professionals, and say, here are the resources that you have. Here's a robot capable of excavation. You can run it yourself right now and try to tackle this challenge. And for professionals and anybody who tries to tackle those challenges, give us your code. And then for NASA engineers,

they can review that and they can see what ideas are out there and they can apply that directly into their missions, into their thinking, or in a lot of cases, bring them on to be part of that as well. This episode is brought to you by Git Prime. Git Prime helps software teams accelerate their velocity and release products faster by turning historical Git data into easy-to-understand insights and reports.

Because past performance predicts future performance, GetPrime can examine your Git data to identify bottlenecks, compare sprints and releases over time, and enable data-driven discussions about engineering and product development. Shift faster because you know more, not because you're rushing. Get started at GetPrime.com slash changelog. That's G-I-T-P-R-I-M-E dot com slash changelog. Again, getprime.com slash changelog. So let's talk about the Easy Razor software project itself and learn about how it all works together. So the primary goal, as it says in the readme of the Easy Razor, is to provide a demonstration robot for visitors at the Kennedy space center. So my first question is, is there a demonstration robot there that the software runs on or is

it a software only demonstration? So there is not currently hardware for the easy razor. When it was first scoped out, that was the goal. And I think in a lot of ways that still is. There's just been no update on the goal. And I think in a lot of ways that still is. There's just been no update on the hardware. But what it does serve because of the simulation that it's able to run, it does serve as a way for you to run it yourself. And so as part of the software package, we have controls built in.

We even have an Android and iOS app to operate the robot. So as soon as the hardware is ready to go, we're going to be able to ship this software on it and get it to operate the robot. So as soon as the hardware is ready to go, we're going to be able to ship this software on it and get it in people's hands, starting at the Kennedy Space Center, because that's easiest. And then hopefully work towards getting it into the hands of more people,

students and universities, et cetera. Yeah, that would be super cool. So what kind of architecture does the software run on? So the entire software for the EasyRazor runs on ROS. ROS is a very popular framework within the robotic community. It stands for Robot Operating System. It is literally the, not just the glue, but the foundation of our entire project. And so in describing ROS, usually it's helpful to explain what it means by an operating software or operating system. It's not like your typical Windows or Linux.

It's really robotic middleware. And so as software engineers, software developers, we use software to control different parts of a robot. And so that can get tricky if you're not used to low-level code and writing drivers and even doing it at scale, right? And so ROS for us in this project provided a way for us to abstract that a little bit and write Python code to operate the robot, but just not directly for hardware. We made our own hardware to be able to test our software on. But the beauty of ROS is that it operates through messages, right? And so we would send a message to move the robot.

And any part of our system that was listening and knew how to respond to it would. And practically, that means if we had hardware listening for a movement message, it would know how to read the message and then it would kick off the drivers and move the wheels. And at the same time, if our simulation was running and it received the message for movement, it would also begin to move. And that was something we didn't have to put an if statement, if hardware do this, if software do that.

We were just able to send a message, move, and any topic capable of receiving that kind of message would enact upon it. And so it made it really easy for us to build the system modularly one component at the time at a time, and also trust just kind of that once we send the message that whoever needed to code up the receiver, you know, that they would just be able to work with it without us having to worry about the intricacies of what they were doing so in what environment did the simulations run so uh ross as a framework comes when you install the full package comes bundled with a simulation software called gazebo and so we initially used ross kinetic which shipped with gazeazebo version seven. And we did all of this inside of Ubuntu. And so as soon as our software spun up, we configured our system, you know, to launch different processes

through a launch file. And so we would kick off the simulation. And immediately, the robot would come onto the screen through the simulation and wait for messages. And so we would either through a gamepad, send messages to it and see it move in real time, or just right there on the command line, be able to interface with it and see it move around, see the drums dig, see the arms go up and down. Yeah, it's interesting. You're talking about controlling this. So what's the, what are some of the control structures for controlling the robot? D-pad, AB, you know, we're talking about Nintendo controllers. We're talking about, you you know what are some of the mechanics for controlling yeah so in terms of a gamepad we stuck to an xbox controller okay we thought it was really natural um in terms of digging to have the

the triggers and the bumpers uh be responsible for forward digging and backwards digging and because they're separated you know visually left and right yeah that's also how we controlled which drum you were trying to rotate and in which direction and then outside of that we implemented a tank turning movement functionality to be able to control the rover and so all that means is you have the two joysticks on it on your gamepad the left joystick would move the left side the left set of wheels either front or back and and the left joystick would move the left side, the left set of wheels, either front or back, and the right joystick would do the same for the right side. And so that was really all we needed. In terms of thinking about controls, it's not too complicated of a robot to operate. You have movement, you have arms going up and down, and then you have digging.

So as you mentioned, you were able to write your quote unquote application code in Python. Was this a language you're already familiar with as you got going and did you learn on the go during this internship how to write Python code? Yeah, so it's funny. I think everybody, there were 10 of us on the project, and I think everybody knew Python really well, except for me. And so that was an interesting thing for me to learn. I had known about Python,

I just hadn't exposed myself to it in terms of using it for a project, which at least being a student in computer science at the time was kind of the anti-pattern, if you will. But thankfully, not too much of a hurdle. ROS itself is very well documented. And so we were able to, if we had an issue, just read the docs and see how it needed to be used. And also, thankfully, we had a lot of smart guys on the team. And so it was very good to be able to lean on kind of some of the more experienced guys. There was one guy on our team, Tiger, who made sure we were PEP8 compliant all the way. He would go through our code and if there was an extra space uh he would he would reject the

pull request and i think that made us better developers for it even if we didn't see it at the time how about the tiger the benevolent dictator there you go gotta get those spaces in there well since we're on a language question i suppose it's worth mentioning your github profile username c sharp ron yes yeah thank you Thank you. Thank you. Uh, that, that pauses is definitely necessary. So I've been playing keyboard for a little over a decade. I mean, not that I'm old or anything, but, uh, C sharp has always been my favorite chord and funny enough, C sharp, the language is the first major language I used for work. And so one day I was at a jam session and my friend Tommy just referred to me as C sharp. He asked me to, to kick off the

song and then the name kind of stuck. And so as part of my SEO strategy, I just put it everywhere I can. I love it. Part of my SEO strategy. You're just trying to make sure nobody calls you McDonald. They're like, call me C sharp. That's the nickname right there. Yeah. I don't know if I need to put pronunciations to nicknames that I give myself because McDonald's is kind of the easy one. It's also, it sticks. I've also been called C chaperone, which, uh, which, which doesn't have the same ring to it no looks like phonetically that's how you would say it but uh sadly i can't use the actual hashtag on a lot of websites because then the browser thinks it's like a id tag so what's where that where'd you get that one from where

that one come from uh c sharp no c sharp ron what'd you say she yeah c sharp ron c sharp ron okay you pronounced it with a accent. I was like, what's that? Oh yeah. Okay. Yeah. That that's the, that's the incorrect saying that I, that I have heard before. If you say it too fast or if there's no capitalization.

That's right. If you lose the camel casing of Ron in there, it's, it's easy to be like, what? C sharp Ron. I get it. Yeah. Yeah. I wasn't sure how, uh, how far I wanted to go with it a few years ago when I was trying to pick my, my GitHub handle. And so I just kind of I get it. Yeah, I wasn't sure how far I wanted to go with it a few years ago when I was trying to pick my GitHub handle.

And so I just kind of went with it. I knew I needed Ron in there because otherwise anybody could be C sharp and I think I'd get a copyright notice. And so it just kind of made sense. I also think C sharp is a ironic nickname for a guy who I happen to know is running Arch Linux right now.

Oh, really? Yeah. I mean, C Sharp is usually somebody who is on Windows or developing for Windows would be using C Sharp as a language. Well, you happen to be on Windows by circumstance because Arch Linux is not the best podcast recording operating system. But is Arch your daily driver? It sure is. When I was first approached about this open source project, I was completely a Microsoft fanboy. I mean, I had been to Ignite and I had Microsoft shirts and stickers. And I found that I was following a lot of Microsoft news, which meant I was always learning,

but it also meant I was kind of closed minded to a lot of the developments in the FOSS community. And so I'm really grateful for this project in particular, because like you said, I am running Arch and I am in this deep rabbit hole of consuming as much open source software as I can and customizing my box. And I think there are a lot of struggles with that. But I think getting, like I mentioned before, getting out of my comfort zone has enabled me to learn a lot more things that I never knew that I would. Like FIM, I think that's made me personally, I'm not going to get into the flame war about it, but I think that's made me personally a better developer. Certainly working with Linux and troubleshooting has gotten me more familiar

with how the operating system functions and having more of that kind of a mindset as a developer, keeping resources, you know, on the front of my mind and not taking things for granted. So you started off in Windows world, see the Old Testament scholars would call that Egypt. You know, you were in the world, and then you've been called out into what we call the wilderness now. You're on your wilderness journey through Arch and Linux, and then eventually you'll come into the promised land, which is macOS. Yes. Actually, one of the guys on the project, Camilo, he's going to hate that I'm going to say this but he has a MacBook Pro

and shortly after he got it it was like the 2018 model, he developed like a mild form of carpal tunnel I don't know if it was the keyboard on it but no quite likely the keyboards are not so hot right now although they just released a brand new one that's supposed to fix

a few things. Yeah, I saw the announcement actually today. It looks really neat. They brought back the escape button. And yeah, I went down that journey like you mentioned and I was very surprised to see how much documentation there is regarding Linux things. I think that helped me a lot. Lots of great resources to learn.

And the only thing that scared me off was looking at old posts from Linus Torvald, comments on requests to change Linux. That really scared me. But yeah, other than that, the experience has been very positive. I ran Ubuntu for a few years during college and shortly thereafter. And back then, this was 2005, 2006. So many things have changed but it was always the wi-fi the wireless card drivers that every time i upgraded i would have to spend hours fixing those things and it just was like i had enough and i know things have gotten a lot

but we still haven't had the year of the linux desktop but i know everybody who runs arch is are usually huge fans of it. I'm a huge fan of Linux. I love that entire I was going to say infrastructure, but what's the word? Just the file system, the commands, the pipes, and all that kind of stuff. I love it all. It's just when it comes time to record a podcast, then you need to use

audio. Adam and I will tend to give a little bit of an eye roll because when someone says they're on Linux, it's like oh, this could be trouble. Maybe that's something i should have said before so you could have steered me steered me away um but it's all good i think it's great and you know ubuntu so the the ross ross development works best on ubuntu and and if you go to the website um i mean the packages are made for the Debian packages. The Ubuntu team now has a robotics team, which is great.

And they're actively working on ROS. And it's really easy, especially with our project, to get it up and running using Ubuntu. You're not missing anything. And I think, frankly, it's the way if you're going to do ROS development to work. Right on. So you accomplished what you said to accomplish in software i love the idea that you have a baseline of features that this first internship you know these 10 students went through and built this foundation and now it can go from there i'm looking at the readme it seems like some of these things the list of bullet points they're on the easy razor readme

might be aspirational at this point i think it would be so cool to get that into the visitor center at kennedy space center so you can like go there and see your software running on a on a real life robot that would be so rad yeah the uh one of the swampworks engineers that we had worked with kurt loykt um he even envisions it envisions it going out to museums you know a big part of the success of nasa going forward, the success of Artemis is going to be kind of public involvement. And to get something like this into as many hands as they can, as many hands as we can, I think is going to be really important. And so if you look at the history of the repo, there are still some commits from, I believe, a month ago. There are a few of us that still work on it out of the 10.

Tiger's listed as the, like you mentioned before, the benevolent dictator for life. It perfectly matched his contribution to the project. And I'm staying on as an admin. And I have work that I intend to do as well. And so we continue to develop even after the deadline, even after we submitted the project and got our A, when so many other teams literally just walked up and left from their projects, we've continued. And we really believe in the project. I think if you ask any of us who have worked on it, we believe in what this project can mean. We believe in,

at least from a robotic standpoint, you know, how much we can learn, how much is still left to be learned about software like this, how much other people can contribute to it that are much smarter than we are certainly. And so it's just, it's been a great ride. And I think 2020 next year is going to be pretty big for our project. We're going to see the next two teams that are working on the easy razor and i can get into specifically what they're working on we're going to get to see their contributions and tiger and i are going to get to review it and uh bring that in and we're really looking forward to that as well is easy razor being used by the artemis mission or is it just part of how you how you got there so artemis i think um for the extent of this is just what I'm working

on, but they're totally separate. The Razor itself that NASA is working on relates to the Mars mission, but I haven't seen where it's slated to be sent out yet. And so it could be part of the Artemis mission. Quite frankly, I don't know. I know that technology like the Razor is going to be important. It's going to be crucial to getting us to mars and staying on there and building a sustainable habitat um in terms of so the the easy razor as a platform is is open source and you can contribute to it if there are really good ideas on there uh breakthroughs in autonomy and and those are even issues that are being worked on now and issues will as ideas come along that we might open up the engineers are looking at that and they they might bring that back in they might reach out for more you know kind of community involvement

and see how stable it is and so this is definitely for education but it has all the possibility of of going further than that. What is up everyone? We are working with Infinite Red to promote their machine learning course. It's called Beginning Machine Learning with TensorFlow.js. Learn more at js.infinite.red. AI used to just be for academics and data scientists, but not anymore. You do not need a PhD, a master's degree, or even a four-year degree to learn about machine learning with TensorFlow.js. All you need is a working knowledge of JavaScript. Artificial intelligence and machine learning is a

fast-growing field that anyone with basic JavaScript knowledge can learn, and the potential upside is so huge. Do yourself a favor and start learning how to apply machine learning to your code today. This is a three-week course covering an introduction to machine learning models, tensors, and the TensorFlow.js framework. The course includes real-world examples, actionable assignments, and quizzes to cement your learning. The course starts now and never ends. Plus, you'll have lifetime access to the course and materials. And of course, because you're a listener, you get a special bonus. You get $100 off. Use the code CHANGELOG when you enroll. Learn more and enroll at

js.infinite.red. And by our friends at Square, we're helping them to announce their new developer YouTube channel. Head to youtube.com slash square dev to learn more and subscribe. Here's a preview of their first episode of the Sandbox Show, where Shannon Skipper and Richard Mute deep dive into the concept of item potency.

Welcome to the pilot episode of The Sandbox Show, a show where we'll... Well, a YouTube show. ...where we'll deep dive into subjects that developers find interesting. Don't worry, there will be plenty of live coding. I'm Shannon, and this is Richard. And we're going to cover a broad range of topics as the show evolves, but for today, what are we going to be covering? On this first episode, we're going to be covering item potency.

We had talked to people in our community, and the thing that people seem to be really confused by is this concept of item potency and how does it relate to interacting with an API. Right. And so I didn't do some Googling on this beforehand, but I know that you did. I did. So the definition of item potency comes from item and potent. So item being same and potent power or potency. So it's the same potency.

All right. Check out this full length show and more on their YouTube channel at youtube.com slash square dev or search for Square Developer. Again, youtube.com slash square dev. Definitely would be a cool way to get started in robotics or even in Python, hopping into this repo and see if you can get it installed, see if you can get it running. The bullet points listed in the readme, the things that it can do, I'm curious how many

of these are aspirational and how many of these are accomplished. So you have Rovercross light to moderate terrain. Collect regolith in rotating drums. Maybe you can explain what some of these things mean. Return regolith to hoppers located away from dig sites. I think I know what that means at least. Execute pre-planned routines. Autonomously navigate around possible obstructions. That sounds like it might be aspirational.

Cooperate in a swarm of other easy razors. That sounds scary, but awesome. How many of these are out there in the code right now? So the biggest one right now is the autonomous movement. If you give it an XY coordinate, you tell it where it needs to go. It will go there and navigate around obstacles in its path. It will dig for a predefined amount of time, I think we have it set to 10 or 15 seconds, and then it will return back to where it started from, which when you spawn it up becomes 0, 0. And so those autonomous functions are working.

Now because it is a simulation, actually digging into the terrain as far as our research went isn't possible in gazebo so to see the terrain change uh would probably need something more robust to be able to like visually see that but for the purposes of a simulation we are able to see it go into the ground and and kind of the whole robot itself moves the hopper that you mentioned there is not one uh there's not a visual model in the simulation. But by returning back to zero zero, we're assuming that that's the base. And so that's a that's accomplished as well. Yeah. And it's funny that you mentioned that. So all of the major highlights are done when it gets more into bringing it to a hopper and bringing regolith in

that's where where hardware would kind of be needed to flesh that out and so regolith itself is an interesting term um so it's the r in easy razor or razor and so that's the top soil in a planet that's your that's your dirt here, your dirt on Mars. And that's what the Easy Razor is trying to collect. They couldn't just call it dirt, huh? They call it regular. Yeah, that's, you know, I don't know if Easy Dazer has the same appeal. That sounds kind of nice.

It's technically a meme. It's a circumference of many things. It includes dust, soil, broken rock, and other related materials. It's more than just his name. Regolith. You know this. I'm just kidding. I don't know this. I don't know any of this. Yeah, you're on dictionary.com over there trying to act smart.

I'm calling myself out on that. That's interesting. The predefined dig times is pretty interesting too. Why, I guess at this point it's meant to be educational and exploratory, so a real mission might be actually having sensors for knowing if you hit bedrock or certain objects that prevent you from actually digging. There might be some more smarts behind the digging process, whereas in this case it's probably a little less smart.

I was going to say dumb, but that's not cool. Less smart, it's just gonna go and dig somewhere at wherever the x y coordinates lead you to for a time collect it and drop it back into a hopper which goes into maybe a train what would you call the thing before the processing plan or yeah like like a home base a central processing plant yeah cool and so that's where swarm comes into play so with the simulation you can easily we feel like we made it easy to just grab your phone or grab a game pad and connect and make it do whatever you want but obviously that doesn't work for real life and the reason

why is if you have a robot on another planet even though we have really fast internet here messages that you send out to other celestial bodies takes forever to get there. And that's if it gets there at all, you know, you got to make sure it's unimpeded. And so you can't just pull up a screen and manual mode, a robot that's on Mars, you have to rely on autonomous communications. And so when you go out to do a dig, you need to determine, you as the robot need to determine, how do I get to my dig site? And is this a viable dig site? Am I going to fall off a cliff somewhere? Is there enough materials to mine here? Or am I going to break my wheels?

And that's where swarm technology really comes into play. Being able to communicate, to have a set of robots communicate with each other and know hey this path i've already determined that you can't cross it uh saves on you know computational resources and possibly even saves the robot from being damaged and by working in a swarm you can also dig more efficiently and so that is probably the the stretchiest of the stretch goals that we had. The implementation we currently have, we're able to spawn up multiple robots and we can give them each a set of XY coordinates to navigate to, dig and come back to. We have that fleshed out and working. And we also have a game version where if you spawn up multiple robots and you hand four people four different phones and or controllers,

they can try to wreck each other. We call that BattleBots. Yeah, that's cool. Yeah, it worked great. So when we demoed BattleBots, which was for our big final presentation in front of the UCF faculty, two of the professors were able to hit each other head on and collision detection in the simulation environment just went haywire. And it became like that scene from a matrix three where Neo and agent Smith are just fighting and like they're

flying in midair as they're fighting. Uh, it, it became a mess and it was so great to see, at least for them. Uh, we were trying to figure out what went wrong in the code. I believe we fixed that since then, but yeah, we're still needing to have that communication layer and that kind of autonomous decision-making to be able to relay back and forth between different robots in the swarm and to be able to truly operate together as a group without any intervention from us or even having us plug in x and y coordinates since we're talking about uh i guess this obstacle

avoidance scenario talk about self-right what was involved in in self-writing is there a lot of interesting things that can come to play there where you know you're upright again is there sensors talk about that a little bit yeah so a really interesting part of the design for the EasyRazor is the fact that it has two arms and two drums that span across to kind of equal the length of the center of the body. And so if you're deployed onto another planet upside down, that's kind of bad because then your readings are all negative or maybe you can't perform certain operations. Because of the structure of the EasyRazor, you can manipulate both arms so that you can flip over. And so flipping over

sounds like a hard thing, but because of the structure of the vehicle is very achievable, especially on other planets where gravity's less most successfully on the moon. But also, so self-right, if you land on your side, if you tip over and land on your side, that sounds almost harder to get out of. But by wiggling your way out and being intelligent about that, you can self-write and make sure you don't fall on your back and so our robot comes with a an imu an inertial measurement unit and as long as the node is spun up the the ross node is spun up to watch out for when those values get skewed we can autonomously self-write and we have at least on our medium post we have a gif of that yeah functioning and we were really impressed because I think Harrison, who's one of the developers, got that working.

And I think he did it within a span of a week. And it's just because we had a stable framework in place. He was able to focus on just the part that he needed to, reading the data, and then moving the arms based on it. Does it actually have an upright or a downright position? It seems it can go either way like it can go either way. It can go either way. We do, in the code, assume that upright is kind of like your base, your positive line, your 0, 0.

It can work upside down, but then all of the math functions would have to be inverted. It's a goofy foot then. Yeah, so it's just best to turn it around. Gotcha. I guess from a software perspective, I guess you could just reverse everything, but you could do it like, you know,

goofy foot mode or something like that. I'm just thinking like, I used to skateboard, so it was like, it was goofy foot for me. Yeah, we could also listen to when that IMU value changes and instead of performing a self-write, if it's safe to do so, just keep going along and invert the values. That would definitely work as well yeah and another interesting part of the design of it is because of the length of the arms and how sturdy the drums

are if you ever needed to you could drive on the drums and we have that working especially on on the moon which is included in our software, where gravity is much less. It's very fun to do and actually very easy. That would be good for if you're going like a descending terrain because you'll have a longer wheelbase, got more stability. Whereas if you've got your drums up, then you've got a shorter wheelbase and you've got potentially, I guess, you'd probably even flip over going down the terrain if you if

you had your drums up like that so yeah the drums themselves are are made to be a lot sturdier than the wheels and so especially if you're going over dangerous terrain there's probably cases where you do want to use the drums and save your wheels because you don't get those back you need to be able to make your way back home until you want to protect your wheels so you couldn't make it self-heal too so that is one of the ideas that was talked about but never implemented if the robot is modular right if you have uh especially like 3d printing uh which i'll get more into in a second uh then you could bring the robot back and just swap out a part um and think about how easy that would be. It makes me think of Wall-E. Yeah, exactly like Wall-E. It would be revolutionary.

But yeah, one of the things that Swampworks is working on is actually this whole idea behind 3D printing. If you need something repaired or you need a utility item, it doesn't make sense to take all of it with you because you don't know what you're going to need, especially if you have a large fleet of hardware. So it's pretty novel and pretty neat to be able to print your own stuff. But then how do you swap out parts? If you have a battery that's dead and you need to swap it out, obviously that can't be 3D printed,

but that's its own kind of autonomy challenge in itself or certainly a visual challenge to not screw up any other part of the robot and change just the part that you need to. But there are videos online of Swampworks and their work towards trying to 3D print parts on other planets based on the regolith. That's already there. Yeah. Let's get some people excited about what you're excited about one of the things you say is your favorite thing to talk about is space and obviously computer science and things like that but uh if you were trying to impress somebody to maybe take a path you've taken or get a lucky chance like you have and i don't mean that lightly

i mean you've obviously worked hard at what you've gotten but uh you even admitted to having to push f5 to refresh to see if someone was playing a joke on you to get this opportunity to go to nasa and do all this cool stuff but if someone was trying to do some things you're trying to do obviously easy razor is probably one way to dig in but what are some paths you can offer to someone who's excited about space computer science like you are yeah so open source um is a a big way to do that i mean there are projects all over the place that are looking for um maintainers and probably not so much within nasa but just in general there are robotics projects out there but i think none of it kind of is possible unless you try to learn something that you don't already know. I fell into the trap for a long while of

only continuing to learn about things that I had already known about until I came across this robotics project. And we live in a great time because there are so many resources to learn basically anything. You just have to decide what you want to learn about. So from the software engineering perspective or in terms of working, I think robotics is going to be a major key in space exploration. There are so many challenges that have yet to be solved, not just here at Kennedy, but in future missions involving, you know, all kinds of research. And so just taking a chance to learn robotics and maybe building your first robot or trying to get ours up and running, just the simulation, I think would be very good. I'm a big believer in kind of seeing the end goal and building a passion around that. And so what I mean is, um, you know,

it's, it's kind of hard to just tell somebody learn robotics and have them want to do it. I think it's much more powerful and captivating to show them like a Rover on, on Mars to show them a real robot and what it can do. Um, and, and, you know, have them pursue their passions that way. About space, nasa.gov is a fantastic website to learn about space. Really kudos to their media team, because every day they post brand new, either images of the day of space or articles about what they're doing, the development with the Artemis mission. It is just, I mean,

it's the landing page, but I think it's, it's kind of the best page to a jumping off point to keep getting excited about space. And if you're a visual person, go to the, go to the centers. You know, you have Johnson Space Center in Houston. If you can physically go to, you have Kennedy Space Center here in Florida. I think nothing, nothing kind of beats being able to see it in person. And yeah, just to kind of bring it back to, you know, to being able to see and know if that's your passion. Definitely check out a center if you can and get other people to go with you and spread the word. The centers are actually really cool. You can see movies there that sort of tell some of the

history you can actually see some of the they always have some things set up because i live here in houston and i we go there at least once a year because we have family coming and we we basically become tourists in our own city whenever we have holidays because we have downtime and family and whatnot and so we're always at the space center doing something and it's so cool to see like what they've done at Mars and all this cool stuff. It's just so wild to see all this history. And I love it even more when you can actually see like the gigantic Saturn

rocket, for example, and just see how extremely big this thing is. Like it's like three football fields long. It's just massive. And just imagine, you know, that in space and what they've done with it.

Yeah. And what's great about it too, is that, you know, those, those kind of remind us of the things that we've done already. There's still so much more left to do. If you could believe it, I actually was kind of turned off to space before this project. I had seen like all the shuttle missions and was excited about the idea, but in terms of actively following it, it just seemed like something very far away or very elusive that I certainly could never be part of. And now, you know, a year later, I see how not true that is.

And not just because I work there. I mean, for literally anybody, NASA has code that they publish. They have tools that they publish themselves that you can download and use. You can see their source code. It's the NASA software catalog. They have projects all over the place to get the public involved and contribute back into their work. And one of them is the Space Robotics Challenge, where registration is open until the end of the year. Anybody can sign up. You just have to be a US citizen, I believe. And you can work on a real robotics project to solve a real challenge that they have based out of Johnson Space Center.

Not that they will make you go there, but it's a remote competition. But that's the team that you're impacting. And I can't stress enough how you don't have to work for NASA to be part of that project. And their community involvement is greater than I've seen in the past, especially because technology is so accessible. They're able to get people who are just excited about space to work on some of their projects. Can we come back to maybe Artemis for a second and talk about maybe where you're currently at? I know that you had mentioned your bachelor's degree in computer science and you're sort of working and going to school at the same time. Maybe touch on that and then maybe touch on specifically some of the things you're

working on with the Artemis mission. Yeah, so I wanted to keep the momentum going as soon as I finished with my bachelor's. So I figured what better to do than go after my master's. A PhD seemed like a lot of work and I'd have to give up work. So I went after my master's. And thankfully, I found a very good program at Georgia Tech that is online, right? So it allows me to continue working full time, like I've been doing since literally I started my bachelor's. And so I'm pursuing that. But then I got a really wonderful opportunity to apply, um, to work as a contractor on the Artemis mission. And I've been working there since September and there are still, so we work on the, on the ground systems equipment. There's the launch control center.

There's one in Johnson, but there's one here in Kennedy as well. Um, and they use a lot of software to measure gas levels to communicate to have failovers it is literally our job to make sure nothing bad happens to the rocket until about 45 seconds into the air and then it's not a problem it goes back to johnson but somebody else's problem at that point not our problem but but if you can if, I mean, those first 45 seconds or however long it's going to be are so crucial. Getting off the ground and getting into space is probably the hardest challenge in addition to re-entry. But once you're in space, you got nothing stopping you.

It's just getting off the ground. And so my team, we support the engineers who are at Kennedy Space Center working on the ground systems. We provide software development support to them, whether that's through automation, whether that's through code reviews and making sure that they're double checking their work from a software perspective, or just, you know, encouraging them to use frameworks so that they're not just, you know, siloed development or making sure that we're not siloed in our development, but that we're able to work towards maintainable code because Artemis, it's just the beginning, right? We, we have such

a long way to go and making the right choices now will help make sure that we can get off the ground to go to the moon in 2024, at least from a software perspective. Do you have to have a security clearance for what you do? I see the red dot about to take me out. No, I do not need security clearance. I do not have security clearance. But there is an extensive background check. There's actually like two background checks I had to go through.

They really vet all of their contractors and all of their employees on center. So that was a fun experience having to go back five years and kind of remember literally everything I did. But yeah, one of my friends who worked on the Easy Razor does have security clearance. And I got to be a character reference for that, which was a really, really fun experience. Yeah. Got that phone call. Hey, so-and-so said you're a reference for that, which was a really, really fun experience. Yeah. Got that phone call. Hey, so-and-so said you're a reference for them. Yeah.

And are they bad or good? Now that you mention it. Digging up that dirt. Yeah, they take security very seriously. Yeah. And regolith, that's all I have to say. But no, if you can imagine now at Kennedy, they've coined the term multi-user spaceport.

And so it's not just NASA. There is SpaceX, like we mentioned, who launches out of there and Boeing and all these other companies. So I think even from that perspective, it's amazing to see, because this is something we haven't seen before, right? Boeing and SpaceX are kind of neck and neck trying to see who's going to be the first

company on their own to send crew back to the ISS, the commercial crew program. And that's like NASA's, I mean, NASA's obviously providing oversight and making sure they're doing it right. But NASA's not dictating how they go about their mission as they would dictate, you know, their own projects. So yeah, it's a really good time to be alive. Even on the Artemis website, they say how important it is for the ground systems that it says, hey, space exploration begins on the ground. So, I mean, hey, that's where we begin, right? You

got to fight the gravity. You got to get that escape velocity going on and boom, you're off. Yeah. And there are so many moving parts to it. I kind of had that one-liner description if we just get it off the ground. Yeah. And I knew it was a difficult task, but actually getting to work with all of the teams, it is an immense task just to do that. And we have the vehicle assembly building,

which has a lot of dedicated men and women working in there and the LCC where the operators are working to make sure that they're ready for the launch for the next Artemis launch. We haven't sent crew into space since I believe 2011. And for the commercial partners, Boeing and SpaceX, they're going to get to do that first. But we're going to get to do that ourselves as a nation truly through this Artemis program. It's interesting, too, that you mentioned that because I'm looking on the site, too, which is very well documented what's going on there for this mission. But it says it begins in 2020. And it's talking about Artemis one being the first

human spacecraft to the moon in the 21st century. Now, I'm not sure if that's tied to the 2020 date line, but it's just sort of giving this overarching, you know, chronology of what's going to happen or what they intend to happen with this mission, which is pretty interesting from Artemis one to, uh, to Artemis three. Yeah. Yeah. And so the goal of the Artemis program is to get us to the moon by 2024. Uh, but we definitely, uh, want to make sure we have the rocketry right before we put people on it. And so Artemis one is going to be so important for that to, to make sure everything goes perfectly and, and make sure we dot our I's and cross our T's to, uh, to know for sure that it's ready for humans. Ready for humans.

That's important. We don't want anybody to die up in space. I remember when that happened. I think I was in second grade when the Explorer blew up on the way, heading up, and it was just such a terrible day. Oh, the Columbia?

Yeah, I don't know if it was Columbia. I can't recall Jared you might remember I was in second grade You were probably in Probably three You don't know this You don't know this

How old is second grade? Uh Six to seven Eight years old Six to seven I don't know Yeah I was I was three

Either way It was a terrible day And we don't want that to repeat I was still eating Lucky Charms That's right That's right Sucking my thumb or something Yeah But it was a terrible day and we don't want that to repeat lucky charms and that's right that's right sucking my thumb or something yeah it was awesome talking to you i mean this has been an amazing i guess journey for you as you mentioned earlier just uh coming from school to open source and that opened up a significant door to nasa and exploring your career further now into

from a bachelor's degree to a master's of science and working at nasa and a mission that may very well change the world because the intention is to go from the moon learn and then be able to take that learning and go to mars which is what a lot of mankind is looking for in terms of you know future space travel and future opportunities to you know do different things there's a lot of opportunity on Mars that is being explored and a lot of hope there for it too. Yeah. And I really want to drive home, you know, it's not just about, I think with NASA specifically, it's not about who, you know,

I think anybody can contribute no matter where you are, no matter where you come from. And now more than ever, we're, we need a lot of that. If we as a nation are going to make this mission a success, and even through the decades, continue to make it a success, it's got to be all hands thing. This isn't some elusive project. This has to come from all of us. And if you check out the Easy Razor project and try to get that up and running, I think it's a good visual way to see certainly one of the projects that are being worked on at NASA and see if you like it, see if you have an interest in robotics,

maybe even check out that space robotics challenge phase two and see if that's something you would like to participate in. It's free to join, free to be a part of. And yeah, the future is bright. I sound really like tagline-ish, but I think that's just because there's a lot of things, and space is giant. The discussions about space are ginormous, but it's not big enough to exclude people. Anybody can be part of it. Is there a link to this challenge?

Because if so, we'll put it in the show notes. Yeah, I think it's the Centennial Challenge. I can certainly send you guys the link to so, we'll put it in the show notes. Yeah, I think it's the Centennial Challenge. I can certainly send you guys the link to it so we can put it in the show notes. We'll do that. I agree. I think space is big, obviously. And I think it's an interesting thing to set your mind on

because there's so many cool things to explore. It's also very dangerous in space. But hey, that's what you dangerous in space, but hey, that's what you got robots for, right? Oh, yeah. Ron, thanks so much for your time today. It was awesome. Thank you both. Alright,

thank you for tuning in to The Change Log. You can discuss this episode on changelog.com. Check the show notes to get started. A huge thanks to our sponsors, and of course, Fastly, Rollbar, and Linode for making everything we do possible. Support our work by telling your friends and leaving us a Check the show notes to get started. you like to change logs, then you will love Practical AI, Brain Science, and everything that we produce. We even have a special podcast called Backstage that you can only get on the Master Feed. Thanks for listening. We'll talk to you next time. Thank you. Субтитры добавил DimaTorzok