Robert Playter: Boston Dynamics CEO on Humanoid and Legged Robotics | Lex Fridman Podcast #374

Robert Playter 0:00 And so our goal was a natural-looking gait. It was real- it was surprisingly hard to get that to work. Um, and we, but we did build a- an early machine. Uh, we called it Petman Prototype. It was the prototype before the Petman robot.

Lex Fridman 0:15 Mm-hmm.

Robert Playter 0:16 And it had a really nice-looking, um, gait where, you know, it would stick the leg out. It would do heel strike first-

Lex Fridman 0:24 Mm-hmm.

Robert Playter 0:24 ... before it rolled onto the toe, so you didn't land with a flat foot. You extended your leg a little bit. Um, but even then, it was hard to get the robot to walk where it wou- when you're walking, that it fully extended its leg. And getting that all to work well took such a long time. In fact, I, I probably didn't really see the nice natural walking that I expected out of our humanoids until maybe last year. And the team was developing on our newer generation of Atlas, you know, some new techniques, um, uh, for developing a walking control algorithm. And they got that natural-looking motion as sort of a byproduct of a, of a just a different process they were applying to developing the control. So that probably took 15 years, uh, 10 to 15 years to sort of get that from, from, you know, the Petman Prototype was probably in 2008 and what was it? 2022. (laughs) Last year that I think I saw good walking on Atlas.

Lex Fridman 1:27 The following is a conversation with Robert Playter, CEO of Boston Dynamics, a legendary robotics company that over 30 years has created some of the most elegant, dextrous and simply amazing robots ever built, including the humanoid robot Atlas and the robot dog Spot, one or both of whom you've probably seen on the internet either dancing, doing back flips, opening doors, or, uh, throwing around heavy objects. Robert has led both the development of Boston Dynamics' humanoid robots and their physics-based simulation software. He has been with the company from the very beginning, including its roots at MIT, where he received his PhD in aeronautical engineering. This was in 1994. At the legendary MIT Leg Lab, he wrote his PhD thesis on robot gymnastics. As part of which, he programmed a bipedal robot to do the world's first 3D robotic somersault. Robert is a great engineer, roboticist and leader. And Boston Dynamics, to me, as a roboticist, is a truly inspiring company. This conversation was a big honor and pleasure, and I hope to do a lot of great work with these robots in the years to come. This is the Lex Fridman podcast. To support it, please check out our sponsors in the description. And now, dear friends, here's Robert Playter.

Lex Fridman 2:57 When did you first fall in love with robotics? (laughs) Let's start with love and robots.

Robert Playter 3:04 Well, love is, is relevant because I think the, the fascination, the deep fascination is really about movement. And, uh, I was visiting MIT looking for a place to get a PhD, and I wanted to do some laboratory work. And, uh, one of my professors at, in the aero department said, "Go see this guy, Marc Raibert, down in the basement of the AI Lab." And so I walked down there and saw him. He showed me his robots, and he showed me this robot doing a somersault.

Lex Fridman 3:37 (laughs)

Robert Playter 3:37 And I just immediately went, "Whoa!" You know?

Lex Fridman 3:40 Yeah.

Robert Playter 3:40 "Robots can do that?" And because of my own interest in, in gymnastics, there was, like this immediate connection. And, um, you know, I was interested in, I was in an a- aero-astro degree because of, you know, flight and movement was all so fascinating to me. And then it turned out that, you know, robotics had this big challenge. How do you, how do you balance? Uh, how do you, how do you build a legged robot that can really get around and that just, that was a fascination. And it still exists today. You're still working on perfecting motion in robots.

Lex Fridman 4:13 What about the elegance and the beauty of the movement itself? Is, is there something maybe grounded in your appreciation of, uh, movement from your gymnastics days? Did you... (sighs) Was there something you just fundamentally appreciated about the elegance and beauty of movement?

Robert Playter 4:31 You know, we had this concept in, in gymnastics of, um, letting your body do what it wanted to do. When you get really good at gymnastics, um, part of what you're doing is putting your, your body into a position where the physics and the body's inertia and momentum will kind of push you in the right direction in a very natural and organic way. And the thing that Marc was doing, you know, in the, um, basement of that laboratory, was trying to figure out how to build machines to take advantage of those ideas. How do you build something so that the physics of the machine just kind of inherently wants to do what it wants to do? And he was building these springy pogo stick type-

Lex Fridman 5:12 Mm-hmm.

Robert Playter 5:13 You know, his first cut at legged locomotion was a pogo stick where it's bouncing and there's a spring mass, uh, system that's oscillating, has its own sort of natural frequency there. And sort of figuring out how to augment those natural physics, um, with also intent. How do you then control that, but not overpower it? It's that coordination that I think creates real potential. We could call it beauty, you know? You could call it, I don't know, synergy?

Lex Fridman 5:41 Mm-hmm.

Robert Playter 5:41 Uh, people have different words for it. Uh, but I think that that was inherent, uh, from the beginning that was clear to me that that, that's part of what Marc was trying to do. He asked me to do that, uh, in my research work. So, um, you know, that's where it got going.

Lex Fridman 5:54 So part of the thing that I think I'm calling elegance and beauty in this case, which was there even with the pogo stick, is maybe, uh, the, the efficiency. So letting the body do what it wants to do.... trying to discover the efficient movement.

Robert Playter 6:06 It's definitely more efficient. It also, um, becomes easier to control in its own way because the, the physics are solving some of the problem itself. It's not like you have to do all this calculation and overpower the physics. The physics naturally, inherently want to do the right thing. Uh, there can even be, you know, uh, feedback mechanisms, stabilizing mechanisms that occur simply by virtue of the physics of the body, and it's, you know, not all, not all in the computer or not even all in your mind as a person. (laughs) And I... There's something interesting in that, that, uh, melding.

Lex Fridman 6:43 You were with Marc for many, many, many years. You were there in this kind of legendary space, uh, of, uh, Leg Lab at MIT in the, (laughs) in, in the basement.

Robert Playter 6:53 (laughs) .

Lex Fridman 6:53 All great things happen in the basement.

Robert Playter 6:55 (laughs) .

Lex Fridman 6:55 Is there some memories... Uh, is there some memories from that time that you have? Because it's so... It's such cutting-edge work in, in, in robotics and artificial intelligence.

Robert Playter 7:09 The memories, the distinctive lessons I would say I, I learned in that, in that time period, and, um, and that I think Marc was a great teacher of, was, uh, it's okay to pursue your interests, your curiosity. Do something because you love it. Um, you'll do it a lot better if you love it.

Lex Fridman 7:30 Mm-hmm.

Robert Playter 7:31 Um, that, that is a, a lasting lesson that I think, uh, we apply at the company still, um, and really is a core value.

Lex Fridman 7:41 So, the interesting thing is I got to, um, uh, with people like, uh, Ross Tedrake and, uh, and others, like, the students that work at those robotics labs are, like, some of the happiest people I've ever met. I don't know what that is (laughs) . I meet a lot of PhD students. A lot of them are kind of broken by the wear and tear-

Robert Playter 8:01 (laughs)

Lex Fridman 8:01 ... or the process. Uh, but roboticists are... While they work extremely hard and work long hours, there's, um, uh, there's a happiness there. The only other group of people I've met like that are people that skydive a lot (laughs) .

Robert Playter 8:16 (laughs) .

Lex Fridman 8:16 Like for s- for some reason, there's a deep fulfilling happiness, maybe from, like, a long period of struggle to get a thing to work, and it works, and there's a magic to it. I don't know exactly 'cause it's so fundamentally hands-on, and you're bringing a thing to life. I don't know what it is, but they're happy.

Robert Playter 8:33 We see... You know, our, our attrition at the company is really low. People come, and they love the pursuit, and I think part of that is that there's perhaps an actual connection to it. It's a little bit easier to connect when you have a robot that's moving around in the world, and part of your goal is to make it move around in the world. You can identify with that. And, and this is on a... This is one of the unique things about the kinds of robots we're building, is this physical interaction lets you perhaps identify with it. So, I think that is a source of happiness. I don't think it's unique to robotics. I think anybody also who is just pursuing something they love, it's easier to work hard at it and be good at it, and, um, uh, m- not everybody gets to find that. Uh, I, I do feel lucky, uh, in that way, and I think, uh, we're lucky as an organization that, that we've been able to build a business around this and that keeps people engaged.

Lex Fridman 9:29 So, if it's all right, let's linger on Marc for a little bit longer, Marc Raibert. So, he, he's a legend. Uh, he's a legendary engineer and roboticist. What, what have you learned about life about robotics from Marc through all the many years you've worked with him?

Robert Playter 9:45 I think the most important lesson, which was, you know, have the courage of your convictions and, and do what you think is interesting. Um, be willing to try to find big, big problems to go after.

Lex Fridman 9:58 Mm-hmm.

Robert Playter 9:59 And at the time, you know, legged locomotion, um, especially in a dynamic machine, w- nobody had solved it, and that felt like a multi-decade problem to go after. And so, you know, have the courage to go after that because you're interested. Uh, don't worry if it's gonna make money. You know, that, that's been, um, a theme. So, that, that's really, uh, probably the most, uh, uh, im- important lesson I think that, uh, I got from Marc.

Lex Fridman 10:28 How crazy is the effort of doing legged robotics at that time especially?

Robert Playter 10:35 You know, Marc got some stuff to work, uh, starting from simple ideas. Uh, uh, so maybe the other... A- another important idea that has really become a value of the company is try to simplify a thing to the core essence. And, and while, you know, Marc was showing videos of animals running across the savanna or, uh, uh, climbing mountains, what he started with was a pogo stick because he was trying to reduce the problem to something that was manageable, and, and, and getting the pogo stick to balance had in it the fundamental problems that if we solve those, you could eventually extrapolate to something that galloped like a horse. And so, look for those simplifying principles.

Lex Fridman 11:19 Um, how, how tough is the job of simplifying a robot?

Robert Playter 11:22 So, I, I'd say in the early days, the, the thing that made Boston... The researchers at Boston Dynamics special is that we, we worked on under- figuring out what that, that central principle was and then building software or machines around that principle, and that was not easy, uh, in the early days. And, and it, it took, um, real expertise in understanding the dynamics of motion and feedback control principles, how to build and... You know, with the computers at the time, how to build a feedback control algorithm that was simple enough that it could run in real time at 1,000 hertz and actually get that machine to work. Um, and that was n- not something everybody was doing, you know, at that time.... now, the world's changing now, and I- I- I think the approaches to controlling robots are going to change. Um, but, uh, uh, and they're going to become more broadly, um, available. Um, but at the time, there weren't many groups who could really sort of work at that principled level, uh, with both the software and- and make the hardware work. I'll- and I'll say one other thing about you're sort of talking about what are the special things. The other thing was go- it's o- it's- it's good to break stuff, you know?

Lex Fridman 12:46 Mm-hmm.

Robert Playter 12:46 Um, you know, use the robots, uh, break them, repair them, um, you know, fix and repeat. (laughs) Test, fix, and repeat. And that- and that's also a core principle that has become part of the company, and it lets you be fearless in your work. Too often, if you are working with a very expensive robot, maybe one that you bought from somebody else or that you don't know how to fix, then you treat it with kid gloves, and you can't actually make progress. You have to be able to break something. And so I think that's, uh, been a- a- a principle as well.

Lex Fridman 13:22 So just to linger on that psychologically. How do you deal with that? 'Cause I remember I had, uh, uh, I built a RC car where it did some, uh, it had some custom stuff like compute on it and all that kind of stuff, cameras, and, uh, because I didn't sleep much, the code I wrote had an issue where it didn't stop the car, and it... The car got confused, and at full speed at like 20, 25 miles an hour slammed into a wall. And I just remember sitting there alone in a deep sadness, um, sort of full of regret, I think, almost anger, um, uh, but also, like, sadness because you think about, "Well, these robots, especially for autonomous vehicles, like, like, you should be taking safety very seriously even in these kinds of things." But just no- no good feelings, um, and made me more afraid probably to do those kind of experiments in the future. Perhaps the right way to have seen that is positively, like is- is to, uh-

Robert Playter 14:25 It depends if you could have built that car or- or just gotten another one, right? That would have been the approach. Um, I remember, um, when I got to grad school, um, you know, I got some training about, uh, operating a lathe and a mill up in the machine shop, and I could start to make my own parts, and I remember breaking some piece of equipment in the lab and then realizing 'cause I... Maybe this was a unique part, and I couldn't go buy it, and I realized, "Oh, I can just go make it." That was an enabling feeling.

Lex Fridman 15:02 Yeah.

Robert Playter 15:02 Then you're not afraid. Yeah, it might take time. It might take more work than you thought it was gonna be required to get this thing done, but you can just go make it, and that's freeing in a way that nothing else is.

Lex Fridman 15:16 You mentioned, um, the feedback control, the dynamics. Sorry for the romantic question, but is... In the early days and even now, is the dynamics, probably more appropriate for the early days, is it more art or science?

Robert Playter 15:33 There's a lot of science around it, and- and trying to develop, you know, scientific principles that let you extrapolate from, like, one-legged machine to another, you know, develop a core set of principles like- like a spring mass bouncing system, and then figure out how to apply that from a one-legged machine to a two or a four-legged machine. Those principles are really important, and- and- and were definitely a core- a core part of our work. Um, there's also, you know, when we started to pursue humanoid robots, um, there was so much complexity in that machine that, you know, one of the benefits of- of the humanoid form is you have some intuition about how it should look-

Lex Fridman 16:20 Mm-hmm.

Robert Playter 16:20 ... while it's moving, and that's a little bit of an art, I think. Now it's... Or maybe it's just tapping into a knowledge that you have deep in your body and then trying to express that in the machine.

Lex Fridman 16:33 Yeah.

Robert Playter 16:33 But that's an intuition that's a little bit more on the art side. Uh, maybe it- it predates your knowledge, you know? Before you have the knowledge of how to control it, you try to work through the art channel. (laughs)

Lex Fridman 16:43 Yeah.

Robert Playter 16:43 And humanoids sort of make that available to you. If it had been a different shape, maybe you wouldn't have had the same intuition about it.

Lex Fridman 16:50 Yeah, so your knowledge about moving through the world is not made explicit to you, so you just... That's why it's art, you get to-

Robert Playter 16:59 And it might... Yeah, it might be hard to actually articulate exactly, You know? (laughs)

Lex Fridman 17:03 Yeah.

Robert Playter 17:03 There's something about... Um, and- and being a competitive, uh, athlete, there's something about seeing a movement. You know, a coach, one of the greatest strengths a coach has is being able to see, you know, some little change in what the athlete is doing and then being able to articulate that to the athlete, you know? And then maybe even trying to say, "And you should try to feel this." Um, so there's something just in seeing, and again, you know, sometimes it's hard to articulate what it is you're seeing, but there's a just perceiving the motion at- at a rate that is, um, uh, again sometimes hard to put into words.

Lex Fridman 17:41 Yeah, I- I wonder, uh, how it is possible to achieve sort of truly elegant movement. You have a movie like Ex Machina. I'm not sure if you've seen it, but, uh, the main actress in that who plays the AI robot, I think is a ballerina. I mean, just the natural, um, elegance and the, I don't know, eloquence of movement is, it's... (laughs) It looks efficient and easy and just... It looks right.... it looks beautiful.

Robert Playter 18:13 It looks right is sort of the key, yeah.

Lex Fridman 18:15 And then you, you look at, um, especially early robots, I mean, they, they, they're so cautious in, in the way they move, um, that it's not, it's not the caution that looks wrong. It's, it's something about the movement that looks wrong, that feels like it's very inefficient, unnecessarily so. And it's hard to put that into words exactly.

Robert Playter 18:38 We think that... And part of the reason why people are attracted to the machines we build is because the inherent dynamics of movement are m- are closer to right.

Lex Fridman 18:48 Mm-hmm.

Robert Playter 18:49 Um, because we, we try to use, you know, walking gaits, or we build a machine around this gait, where you're trying to work with the dynamics of the machine, instead of to stop them.

Lex Fridman 19:01 Mm-hmm.

Robert Playter 19:01 You know, some of the early walking machines, you know, you're essentially... You're really trying hard to not let them fall over, and so you're always stopping the tipping motion, you know?

Lex Fridman 19:12 Yeah.

Robert Playter 19:12 And sort of the insight of dynamic stability in a legged machine is to go with it, you know? (laughs) Let the tipping happen, you know, let yourself fall, but then catch your, catch yourself-

Lex Fridman 19:23 Mm-hmm.

Robert Playter 19:23 ... with that next foot. And there's something about getting those physics to be expressed in the machine that people interpret as life-like-

Lex Fridman 19:34 Mm-hmm.

Robert Playter 19:34 ... or, or elegant, or just natural looking. And so I think if you get the physics right, it also ends up being more efficient, likely. There's a benefit that it probably ends up being more stable in the long run. You know, it could, it could walk stably over a wider r- range of conditions. Um, and it's, uh, and it's more beautiful and attractive at the same time.

Lex Fridman 19:58 So,

Lex Fridman 19:59 how hard is it to get the humanoid robot Atlas to do some of the things it's recently been doing? Let's forget the flips and all of that. Let's just look at the running. Maybe you can correct me, but there's something about running... I mean, that's not careful at all. That's, you're falling forward. You're jumping forward and are falling, so how h- hard (laughs) is it to get that right?

Robert Playter 20:20 Our first humanoid, we needed to deliver natural looking walking. You know, uh, we took a contract, uh, from the Army. They wanted a robot that could, uh, walk naturally. They wanted to put a suit on the robot and be able to test it in a gas e- environment. And so they wanted the na- the, the motion to be natural. Um, and so our goal was a natural looking gait. It was real... It was surprisingly hard to get that to work. Um, and we... But we did build a, an early machine, uh, we called it Petman Prototype. It was the prototype before the Petman robot.

Lex Fridman 20:56 Mm-hmm.

Robert Playter 20:57 And it had a really nice looking, um, gait, where, you know, it would stick the leg out. It would do heel strike first-

Lex Fridman 21:05 Mm-hmm.

Robert Playter 21:05 ... before it rolled onto the toe, so you didn't land with a flat foot. You extended your leg a little bit, um, but even then, it was hard to get the robot to walk where it would when you're walking, that it fully extended its leg, and, and essentially landed on an extended leg. And if you watch closely how you walk, you probably land on an extended leg, but then you immediately flex your knee as you start to make that contact.

Lex Fridman 21:28 Mm-hmm.

Robert Playter 21:28 And getting that all to work well took such a long time. In fact, I, I probably didn't really see the nice natural walking that I expected out of our humanoids until maybe last year. And the team was developing on our newer generation of Atlas, you know, some new techniques, um, uh, for developing a walking control algorithm, and they got that natural looking motion as sort of a byproduct of a, of a just a different process they were applying to developing the control. So, that probably took 15 years, uh, 10 to 15 years to sort of get that from, from, you know... The Petman Prototype was probably in 2008, and what was it? 2022, (laughs) last year that I think I saw good walking on Atlas.

Lex Fridman 22:16 I- if you could just, like, linger on it, what are some challenges of getting good walking? So, is it, um, is this, (laughs) is this partially like a hardware, like, actuator problem? Is it the control? Is it the artistic element of just observing the whole system operating in different conditions together? I mean, is there some kind of interesting quirks or challenges you can speak to, like the heel strike or all this kind of-

Robert Playter 22:41 Yeah, so one of the things that makes the, like, this straight leg, uh, a challenge is you're sort of up against a, a singularity, a m- a mathematical single- singularity, where, you know, when, when your leg is fully extended, it can't go further the other direction, right?

Lex Fridman 22:56 Mm-hmm.

Robert Playter 22:56 There's only... You can only move in one direction, and that makes all of the calculations around how to produce torques at that joint, or positions, makes it more complicated. And so having all of the mathematics so it can deal with these singular configurations is one of many (laughs) challenges, uh, uh, that we face. And, and I'd say in, in the, you know, in those earlier days, again, we were working with these really simplified models, so we're trying to boil all the physics of the complex human body into a simpler subsystem that we can more easily describe in mathematics. And sometimes those simpler subsystems don't have all of that complexity of the s- straight leg built into them. And so, um, what's, what's happened more recently is we're able to apply techniques that let us take the full physics of the, um, uh, robot into account, and, and deal with some of those, uh, strange situations like the, like the straight leg.

Lex Fridman 23:59 So, is there a fundamental challenge here that it's... Uh, maybe you can correct me, but is it underactuated? Are you falling?

Robert Playter 24:06 Underactuated is, is the right word, right? You can't, you can't, uh, push the robot in any direction you want to.

Lex Fridman 24:12 Yeah.

Robert Playter 24:12 Right? And so that, that is one of the hard problems of, of, uh, legged locomotion.

Lex Fridman 24:16 And you have to do that for natural movement?

Robert Playter 24:19 It's not necessarily required for natural movement. It's just required... You know, we, we don't have, you know, a gravity force that you can hook yourself onto to apply, uh, an, an external force in the direction you want at all times, right? The only-

Lex Fridman 24:34 Yeah.

Robert Playter 24:34 ... the only external forces are being mediated through your feet, and how they get mediated depend on how you place your feet. And, uh, you know, you can't just, uh, you know... God's hand can't reach down and give and, and push in any direction you want, (laughs) you know, so...

Lex Fridman 24:49 Is there, uh, is there some extra challenge to the fact that Atlas is such a big robot?

Robert Playter 24:54 There is. The humanoid form is, um, um, attractive in many ways, but it's also a challenge in many ways. Um, y- you have this big upper body that has a lot of mass and inertia, um, and throwing that inertia around increases the complexity of maintaining balance. And as soon as you pick up something heavy in your arms, you've made that problem even harder. And so, uh, in the early work, in the leg lab and in the early days at the company, you know, we were pursuing these quadruped robots, which had a, a kind of built-in simplification. You had this big rigid body and then really light legs. So, when you swing the legs, the leg motion didn't impact the body motion very much. All the mass and inertia was in the body. But when you have the humanoid, that doesn't work. You have big heavy legs. You swing the legs, it affects everything else. And so dealing with all of that interaction does make the humanoid a much more complicated platform.

Lex Fridman 25:58 And I also saw that, uh, at least recently, you've been doing more explicit modeling of the stuff you pick up.

Robert Playter 26:04 Yeah. Yeah.

Lex Fridman 26:05 Which is very r- (laughs) um, really interesting. So, you have to, what? Model the shape, the weight distribution? I don't know. What... Like, you have to under- like, include that as part of the modeling, as part of the planning 'cause... Okay, so for people who don't know, uh, so Atlas, at least in, like, the recent video, like, throws a heavy bag, throws a bunch of-

Robert Playter 26:30 Yeah.

Lex Fridman 26:30 (laughs) ... stuff. So, what, what's involved in, uh, picking up a thing, a heavy thing, uh, and when that thing is a bunch of different nonstandard things? I think it also picked up, like, a barbell and, uh, to be able to throw it in some cases. What's, what are some interesting challenges there?

Robert Playter 26:48 So, we were definitely trying to show that the robot and the techniques we're applying to the ro- uh, to Atlas let us deal with heavy things in the world.

Lex Fridman 26:58 Yeah.

Robert Playter 26:58 Because if the robot's going to be useful, it's actually got to move stuff around.

Lex Fridman 27:01 Yeah.

Robert Playter 27:02 And that, and that needs to be significant stuff that's a- an appreciable portion of the, the body weight of the robot. And we also think this differentiates us from the other humanoid robot activities that you're seeing out there. Mostly, they're not picking stuff up yet, uh, not heavy stuff anyway. Um, but just like you or me, you know, you need to anticipate that moment, you know? You're reaching out to pick something up, and as soon as you pick it up, your center of mass is gonna shift, and if you're gonna, you know, turn in a circle, you have to take that inertia into account. And if you're gonna throw a thing, you know, you've got... A- all of that has to be sort of included in, in the model of what you're trying to do. So, the robot needs to have some idea or expectation of what that weight is and then, and sort of predict, you know, think a couple of seconds ahead, "How do I manage my o- now my, my body plus this big heavy thing together (laughs) to get, and, and still maintain balance?" Right? And so, um, I, I, uh... That's a big change for us, and I think the tools we've built are really allowing that to happen, um, quickly now. Some of those motions that you saw in that most recent video, we were able to create in a matter of days. It used to be that it took six months to do anything new, you know-

Lex Fridman 28:18 Yeah.

Robert Playter 28:18 ... on the robot, and, and now we're starting to develop the tools that let us do that in a matter of days. And so, we think that's really exciting. That means that the ability to create new behaviors for the robot is gonna be a, um, a quicker process.

Lex Fridman 28:31 So, being able to explicitly model new things that it might need to pick up, new types of things.

Robert Playter 28:37 And, you know, to some degree, you don't, you don't want to have to pay too much attention to each specific thing, right? Um, there's sort of a generalization here. Uh-

Lex Fridman 28:48 Yeah.

Robert Playter 28:48 ... um, obviously when you grab a thing, you have to conform your, your hand, your end effector to the surface of that shape. But once it's in your hands, it's probably just the mass and inertia that matter, and the, the shape may, may not be as important.

Lex Fridman 29:03 Yeah.

Robert Playter 29:03 And so, you know, for some, in some ways, you want to pay attention to that detailed shape. In, in others, you want to generalize it and say, "Uh, well, uh, all I really care about is the center of mass of this thing, especially if I'm going to throw it up on that scaffolding."

Lex Fridman 29:17 And it's easier if the body is rigid. What if it's, there's some... Doesn't it throw, like, a sandbag type thing?

Robert Playter 29:22 That tool bag, you know, had-

Lex Fridman 29:24 Tool bag.

Robert Playter 29:24 ... loose, had loose stuff in it.

Lex Fridman 29:25 Yeah.

Robert Playter 29:26 So, it, it, it, it managed that. There are harder things that we haven't done yet. You know, we could have had a big jointed thing, or-

Lex Fridman 29:33 Yeah.

Robert Playter 29:33 ... I don't know, a bunch of loose wire or rope.

Lex Fridman 29:35 What about carrying another robot? How about that? (laughs)

Robert Playter 29:39 (laughs) Yeah, we haven't, we haven't done that yet. Although-

Lex Fridman 29:41 Carry a spot.

Robert Playter 29:41 ... I guess we did a little bit of a... We did a, uh, a little skit around Christmas where we had-

Lex Fridman 29:47 Yeah.

Robert Playter 29:47 ... two spots holding up another spot that was trying to put, you know, a bow on a tree. So-

Lex Fridman 29:51 That's good.

Robert Playter 29:51 ... I guess we're doing that in a small way. (laughs)

Lex Fridman 29:54 Okay, that's pretty good. Uh, let me ask the all-important question. Uh, do you know how much Atlas can curl against a barbell?

Robert Playter 30:02 (laughs)

Lex Fridman 30:02 Have you... (laughs) I mean, you know, this, for us humans, that's really one of the most fundamental questions you can ask another human being.

Robert Playter 30:11 (laughs)

Lex Fridman 30:13 Curl, bench, deadlifts?

Robert Playter 30:14 It probably can't curl as much as we can yet, but, uh, a metric that I think is interesting-

Lex Fridman 30:19 Yeah.

Robert Playter 30:20 ... is, um, you know, another way of, of looking at that strength-... is, you know, the box jump. So if, how high of a box can you jump onto?

Lex Fridman 30:29 Good question.

Robert Playter 30:30 And, uh, Atlas, I don't know the exact height. It was probably a meter high or something like that. It was a pretty, pretty tall jump that Atlas was able to manage, uh, when we last tried to do this. And, and I have video of, uh, my chief technical officer doing the same jump, and he really struggled, you know, to get up.

Lex Fridman 30:47 Oh, the human?

Robert Playter 30:47 The human, getting all the way on top of this box.

Lex Fridman 30:50 Mm-hmm.

Robert Playter 30:50 But then, you know, Atlas w- was able to do it. Um, we're now thinking about the next generation of Atlas, and we're probably gonna be in the realm of a person can't do it, you know, with this, with the next generation. You know, the, the robots, the actuators are gonna get stronger, where it really is the case that at least some of these joints, some of these motions will be stronger.

Lex Fridman 31:10 And to understand how high it can jump, you probably had to do quite a bit of testing.

Robert Playter 31:14 Oh, yeah. And there's lots of videos of it trying and failing, and that's, you know, that's all, you know. We don't always release those, those videos, but they're a lot of fun to look at. (laughs)

Lex Fridman 31:23 Uh, so, we'll, we'll talk a little bit about that. Uh, but if we c- can you talk to the jumping? 'Cause you talked about the walking, and it took a long time, many, many years to get the walking to be natural. But there's also really natural-looking, uh, robust, resilient jumping. How hard is it to do the jumping?

Robert Playter 31:43 Well, again, this stuff has really evolved rapidly in the last few years. You know, the first time we did a somersault, um, you know, there was a lot of kinda manual iteration. What is the trajectory? You know, how hard do you throw y- In fact, uh, in these early days, uh, I actually would, when I'd see early experiments that the team was doing, I might make suggestions about how to change the technique, again, kind of borrowing from my own intuition about how backflips work. Um, but frankly, they don't need that anymore. So, in the, in the early days, you had to iterate kind of in almost a manual way, trying to change these trajectories of the arms or the legs-

Lex Fridman 32:22 Mm-hmm.

Robert Playter 32:22 ... uh, to try to get th- you know, a successful backflip to happen. Um, but more recently, we're running these model predictive, uh, uh, control techniques, where we're able to... The robot essentially can think in advance for the next second or two about how its motion is going to transpire.

Lex Fridman 32:43 Mm-hmm.

Robert Playter 32:43 And you can, you know, solve for optimal trajectories to get from A to B. So, this is happening in a much more natural way, and, and we're really seeing an acceleration happen in the development of these behaviors, again, partly due to these, um, optimization techniques, uh, sometimes learning techniques. Um, so it's, there's, it's hard, in that there's a lot of mathematics in, uh, behind it. Uh, but we're figuring that out.

Lex Fridman 33:12 So, you can do model predictive control for, uh... I mean, I don't even understand what that looks like when the entire robot is in the air flying and doing a backflip. (laughs)

Robert Playter 33:22 Yeah, well-

Lex Fridman 33:23 I mean... (laughs)

Robert Playter 33:24 But, but, but that's the cool part, right? So, you know, you-

Lex Fridman 33:27 Yeah.

Robert Playter 33:27 You know, the physics, we, we can calculate physics pretty well using, you know, Newton's laws about how it's going to evolve over time and the ro- You know, this, this, the sick trick, which was a s- front somersault with a half twist-

Lex Fridman 33:39 Yeah.

Robert Playter 33:39 ... is a good example, right? You saw the robot on various f- versions of that trick, I've seen it, land in different configurations, and it still manages to stabilize itself. And so, you know, what this model predictive control means is, again, the, in real time, the robot is projecting ahead, you know, a second into the future and sort of exploring options. "And if I, if I move my arm a little bit more this way, how is that gonna affect the outcome?" And so it can do these calculations, many of them, you know, uh, and, and basically solve for, you know, "Given where I am now, maybe I took off a little bit screwy from how I had planned. I can adjust."

Lex Fridman 34:21 So, you're adjusting in the air for the leg-

Robert Playter 34:23 Adjust on the fly. So, the, the model predictive control lets you adjust on the fly. And of course, I think this is what, you know, people adapt as well. We, we, when, when we do a, even a gymnastics trick, we try to set it up so it's as close to the same every time. But we figured out how to do some adjustment on the fly. And now, we're starting to figure out that the robots can do this adjustment on the fly as well, using these techniques.

Lex Fridman 34:46 In the air. That's so... I mean, it just feels, from a robotics perspective, just surreal.

Robert Playter 34:53 Well, that's sort of the... Uh, you talked about under-actuated, right?

Lex Fridman 34:55 Yeah.

Robert Playter 34:55 So when you're-

Lex Fridman 34:56 That's really... (laughs)

Robert Playter 34:56 When, when you're in the air, there's something, there's some things you can't change, right? You can't-

Lex Fridman 35:01 Yeah.

Robert Playter 35:01 ... change the momentum while it's in the air 'cause you can't apply an external force or torque. And so, the momentum isn't gonna change. So, how do you work within the constraint of that fixed momentum to still get from A to B, (laughs) where, where you wanna be?

Lex Fridman 35:14 That's really under-actuated. (laughs)

Robert Playter 35:16 (laughs)

Lex Fridman 35:17 You're in the air. I mean, you bec- you become a drone for a brief moment in time. No, you're not even a drone 'cause you can't- (laughs)

Robert Playter 35:23 Can't hover.

Lex Fridman 35:24 You can't hover. You can't-

Robert Playter 35:26 You're gonna, you're gonna impact soon. Be ready. (laughs)

Lex Fridman 35:28 Yeah. Have you considered, uh, like a hover type thing, or no? No.

Robert Playter 35:31 No.

Lex Fridman 35:31 It adds too much weight.

Robert Playter 35:33 No.

Lex Fridman 35:33 (laughs) I mean, it's just, it's just incredible. Uh, and just even to have the guts to try a backflip with such a large body, that's wild.

Robert Playter 35:43 (laughs)

Lex Fridman 35:43 What, like, uh, how-

Robert Playter 35:44 Well, we definitely broke a few robots trying.

Lex Fridman 35:46 (laughs)

Robert Playter 35:46 (laughs) But that, but that's where the build it, break it, fix it, you know, uh, strategy comes in. You gotta be willing to break... And what ends up happening is you end up, by breaking the robot repeatedly, you find the weak points, and then you end up redesigning it, so it doesn't break so easily next time.

Lex Fridman 36:01 Yeah.

Robert Playter 36:01 You know? (laughs)

Lex Fridman 36:02 Through the breaking process, you learn a lot, like a lot of lessons, and you keep improving not just how to make the backflip work, but everything.

Robert Playter 36:10 Yeah, and how to build a machine better.

Lex Fridman 36:11 Yeah.

Robert Playter 36:12 Yeah.

Lex Fridman 36:13 I mean, is there something about just the guts to come up with an idea of saying, "You know what? Let's try to make it do a backflip"?

Robert Playter 36:21 Well, I think the courage to do a backflip in the first place and, and to not worry too much about the ridicule of somebody saying, "Why the heck are you doing backflips with robots?"

Lex Fridman 36:30 Sure.

Robert Playter 36:31 Because a lot of people have asked that, you know, "Why-"

Lex Fridman 36:33 (laughs)

Robert Playter 36:34 ... why, (laughs) why are you doing this?

Lex Fridman 36:35 Why go to the moon-

Robert Playter 36:36 (laughs)

Lex Fridman 36:37 ... in this decade and do the other things, JFK?

Robert Playter 36:40 (laughs)

Lex Fridman 36:41 Not because it's easy. Because it's hard.

Robert Playter 36:43 Yeah, exactly. (laughs)

Lex Fridman 36:46 (laughs) Don't ask questions. Okay, so the j- uh, the jumping. I mean, it's just th- there's a lot of incredible stuff. If we can just rewind a little

Lex Fridman 36:53 bit to, uh, the DARPA Robotics Challenge in 2015, I think, which was, for people who aren't familiar with the DARPA challenges, it, uh, was f- first with autonomous vehicles, and there's a lot of interesting challenges around that. And the DARPA Robotics Challenge was when, uh, humanoid robots were tasked to do all kinds of, uh, uh, you know, manipulation, walking, uh-

Robert Playter 37:21 Driving a vehicle.

Lex Fridman 37:21 ... driving a car, all these kinds of challenges, with, if I remember correctly, sort of some slight capability to communicate with humans. But, uh, the communication was very poor, so basically it has to be almost entirely autonomous.

Robert Playter 37:37 It could have periods where the communication was entirely interrupted and the robot had to be able to proceed.

Lex Fridman 37:42 Yeah.

Robert Playter 37:42 But you could provide some high-level guidance to the robot, basically low, low-bandwidth communications, uh-

Lex Fridman 37:48 Yeah.

Robert Playter 37:48 ... to steer it.

Lex Fridman 37:49 I watched that challenge with kind of tears in my eyes, eating popcorn. W- with har-

Robert Playter 37:55 Us too. (laughs)

Lex Fridman 37:56 (laughs) But I w- I wasn't personally losing, uh, you know, uh, hundreds of thousands, millions of dollars and many years of incredible hard work by some of the most brilliant roboticists in the world. So that, that was why the tragic, that's why-

Robert Playter 38:10 (laughs)

Lex Fridman 38:10 ... the tears came. So anyway, what, what have you, um... Just looking back to that time, what have you learned from that experience? Uh, maybe if you could describe what it was, uh, sort of to set up for people who haven't seen it.

Robert Playter 38:23 Well, so there was a contest where a bunch of different, um, robots were asked to do a series of tasks, uh, some of those that you mentioned, drive a vehicle, get out, open a door, go identify a valve, shut a valve, use a tool to maybe cut a hole in, um, a, a surface, and then crawl over some stairs and maybe some rough terrain. So, it was, the idea was have a, a general-purpose robot that could do lots of different things. Um, had to be mobility and manipulation, onboard perception. And there was a contest, uh, which DARPA likes, uh, at the time was running sort of follow on to the, the Grand Challenge, which was, let's, let's try to push vehicle autonomy along, right? They, they, they encouraged people to build autonomous cars. So, they're trying to basically push an industry forward.

Lex Fridman 39:20 Mm-hmm.

Robert Playter 39:21 And, um, uh, we were asked... Our role in this was to build, um, a humanoid. At the time, it was our sort of first generation Atlas robot. And, uh, we built maybe 10 of them. I don't remember the exact number. Uh, and DARPA distributed those to various teams, um, that sort of won a, a contest, uh, showed that they, uh, could, you know, program these robots and then used them to compete against each other. And then, other robots were introduced as well. Some teams built their own robots. Carnegie, um, Mellon, for example, built their own robot. And, uh, and all these robots competed to see who could sort of get through this, this maze, um, uh, the fastest. And, uh, again, I think the purpose was to kind of push the whole industry forward. Uh, we provided the robot and some baseline software, but we didn't, we didn't actually compete as a participant-

Lex Fridman 40:19 Mm-hmm.

Robert Playter 40:20 ... uh, where we were trying to, uh, you know, drive the robot through this maze. Uh, we were just trying to support the other teams. It was humbling because it was, it was really a hard task. And, and honestly, the robots, the tears were because mostly the robots didn't do it. (laughs) You know?

Lex Fridman 40:36 Yeah.

Robert Playter 40:36 They fell down, you know, uh, repeatedly. Um, it was hard to get through this contest. Uh, you know, some did, and, and, you know, they were rewarded and won. But it was humbling because of just how hard... These tasks weren't all that hard. A person could have done it very easily. Um, but it was really hard, uh, to get the robots to do it, you know? And, uh-

Lex Fridman 40:57 Well, the-

Robert Playter 40:57 ... hard-

Lex Fridman 40:57 ... the general nature of it, the, the variety of it.

Robert Playter 41:00 The variety.

Lex Fridman 41:01 And also, that I don't know if the tasks were, (sighs) sort of... The task in themselves help us understand what is difficult and what is not. I don't know if that was obvious before the contest was designed. So, you kind of tried to figure that out. And I think, uh, Atlas is really a general robot platform, and it's perhaps not best suited for the specific tasks of that contest. Like, for, for just for example, probably the hardest task is not the driving of the car, but getting in and out of the car.

Robert Playter 41:35 (laughs)

Lex Fridman 41:35 And Atlas probably... You know, if you were to design a robot that can get into the car easily and get out easily, you probably would not make Atlas, that particular car.

Robert Playter 41:45 Yeah. The, the robot was a little bit big-

Lex Fridman 41:47 Yeah.

Robert Playter 41:47 ... to get in and out of that car.

Lex Fridman 41:49 (laughs)

Robert Playter 41:49 Right?

Lex Fridman 41:49 It, it doesn't fit, yeah.

Robert Playter 41:50 But th- this is the curse of a general-purpose robot, that they're not perfect at any one thing, uh, but they might be able to do a wide variety of things. And, and that is, that is the goal at the end of the day. You know, I think we all wanna build general-purpose robots that can be used for lots of different activities, but it's hard. And, um, a- and the wisdom in, in building successful robots up until this point have been, go build a robot for a specific task, and it'll do it very well.

Lex Fridman 42:23 Mm-hmm.

Robert Playter 42:24 And as long as you control that environment, it'll operate perfectly. But, but robots need to be able to deal with uncertainty. If they're gonna be useful to us in the future, they need to be able to deal with unexpected, uh, situations. And that's sort of the goal of a general-purpose or multi-purpose robot.And that's just darn hard. And so, some of... You know, there are these curious little failures, like I remember one of the... A robot, you know, the first, the first time you start to try to push on the world with a robot-

Lex Fridman 42:54 Mm-hmm.

Robert Playter 42:55 ... you, you forget that the world pushes back and, and will push you over (laughs) if you're not ready for it. And the robot, you know, reached to grab the door handle. I think it missed the grasp of the door handle.

Lex Fridman 43:07 Mm-hmm.

Robert Playter 43:08 Was expecting that its hand was on the door handle.

Lex Fridman 43:10 Mm-hmm.

Robert Playter 43:10 And so, when it tried to turn the knob, it just threw itself over. It didn't realize, "Oh, I had missed the door handle. I didn't ha-

Lex Fridman 43:17 Mm-hmm.

Robert Playter 43:18 "I didn't... I was expecting a force back from the door. It wasn't there, and then I lost my balance." So, these little simple things that you and I would take totally for granted and deal with (laughs) the robots-

Lex Fridman 43:29 Mm-hmm.

Robert Playter 43:30 ... don't know how to deal with yet. And so, you have to start to deal with all of those, uh, circumstances. (laughs)

Lex Fridman 43:35 Well, I think a lot of us experience this in, uh, even when sober, but drunk too. Uh, sort of you pick up a thing and expect it to be, what is it, heavy, and it turns out to be light.

Robert Playter 43:49 Yeah, and then whoa.

Lex Fridman 43:49 Oh, yeah.

Robert Playter 43:50 Yeah.

Lex Fridman 43:50 And then, so the same... And I'm sure if your depth perception, for whatever reason, is screwed up, if you're, if you're drunk or some other reason, and you think you're putting your hand on the table, and you miss it. I mean, it's the same kind of situation.

Robert Playter 44:03 Yeah.

Lex Fridman 44:04 Uh, but there's a-

Robert Playter 44:05 Which is why you need to be able to predict forward just a little bit. And so, that's where this model of predictive control stuff comes in.

Lex Fridman 44:11 Ah.

Robert Playter 44:11 Predict forward what you think is gonna happen, and then if, and if that does happen, you're in good shape. If something else happens, you better start predicting again quickly. (laughs)

Lex Fridman 44:19 So, re- re- like, re- re- uh, regenerate a plan-

Robert Playter 44:23 Yeah.

Lex Fridman 44:23 ... when you don't... I mean, that, um, that also requires a very, uh, fast feedback loop of updating, uh, what your prediction, how it matches to the actual real world.

Robert Playter 44:36 Yeah, those things have to run pretty quickly.

Lex Fridman 44:38 What's the challenge of running things pretty quickly, 1,000 hertz of acting and sensing quickly?

Robert Playter 44:47 You know, there's a few different layers of that. You, you want, at the lowest level, you like to run things typically at around 1,000 hertz, which means that, you know, at each joint of the robot, you're measuring position or force and then trying to control your actuator, whether it's a hydraulic or electric motor, trying to control the force coming out of that actuator. And you wanna do that really fast, something like 1,000 hertz. And that means you can't have too much calculation going on at that joint. Um, but that's pretty manageable these days, and- and it's fairly common. And then, there's another layer that you're probably calculating, you know, maybe at 100 hertz, maybe 10 times slower, which is now starting to look at the overall body motion and thinking about the, the larger physics of, of the, um, of the robot. Um, and then there's yet another loop that's probably happening a little bit slower, which is where you start to bring, you know, your perception in, your vision, and things like that. And so, you need to run all of these loops sort of simultaneously. You do have to manage your, your computer time so that you can squeeze in all the calculations you need, uh, in real time in a very consistent way. Um, and the amount of calculation we can do is increasing as computers get better, which means we can start to do more sophisticated calculations. I can have a more complex model doing my forward prediction, and- and that might allow me to do even better predictions as I, as I get better and better. And- and it used to be... Again, we ha- you know, 10 years ago, we had to have pretty simple models that we were running, you know, at those fast rates 'cause the computers weren't as capable about calculating forward with a, a sophisticated model. But as- as computation gets better, we can- we can do more of that.

Lex Fridman 46:42 What about the actual pipeline of software engineering? How easy it is to keep updating Atlas, like, do continuous development on it? So, how many computers are on there? Is there a nice pipeline?

Robert Playter 46:56 It's an important part of, um, building a team, um, around it, which means, you know, um, you need to also have software tools, simulation tools, you know? So, um, we have always made strong use of, uh, physics-based simulation tools to do, uh, some of this calculation, basically test it in simulation before you put it on the robot. But you also want the same code that you're running in simulation to be the same code you're running on the hardware. And so, even getting to the point where it was the same code going from one to the other, we probably didn't really get that working until, you know, a few year- several years ago.

Lex Fridman 47:36 Mm-hmm.

Robert Playter 47:37 Um, but that was a... You know, that was a bit of a milestone. And so, you wanna work- certainly work these pipelines so that you can make it as easy as possible and have a bunch of people working in parallel, especially when you... You know, we only have, you know, four of the Atlas r- robots, the modern, uh, Atlas robots at the company. And, you know, we probably have, you know, 40 developers there all trying to gain access to it. And so, you need to share resources and use some of these, uh, some of this software pipeline.

Lex Fridman 48:04 Well- well, that's a really exciting step to be able to run the exact same code and simulation as on the actual robot. Uh, how hard is it to do, uh, realistic simulation, physics-based simulation of- of- of Atlas such that... I mean, the- the dream is like if it works in simulation, it works perfectly in reality.

Robert Playter 48:25 (laughs)

Lex Fridman 48:25 How hard is it to sort of clo- keep working on closing that gap?

Robert Playter 48:29 The root of some of our physics-based simulation tools really started at MIT, and, um, we built some, some, some good physics-based modeling tools there. The early days of the company, we were trying to develop those tools as a commercial product, so we continued to develop them.... it wasn't a particularly successful commercial product. But we ended up with some nice physics-based simulation tools so that when we started doing legged robotics again, we had a really nice tool to work with. And the things we paid attention to were, were things that weren't necessarily handled very well in the commercial tools you could buy off the shelf, like-

Lex Fridman 49:03 Mm-hmm.

Robert Playter 49:03 ... like interaction with the world, like foot ground contact. And so, trying to model those contact, um, events well in a way that captured the important parts of, of the interaction wa- was a really important element, uh, to get right, and to also do in a way that was computationally feasible-

Lex Fridman 49:26 Mm-hmm.

Robert Playter 49:26 ... um, a- and could run fast. 'Cause if you, if your simulation runs too slow, you know, then your developers are sitting around waiting for stuff to run and compile. So, it, it's always about efficient, uh, uh, fast operation as well. So, that's been a big part of it. You know, I think developing those tools in parallel to the development of the, of the platform and trying to scale them has, has really been, uh, essential, I'd say, to us being able to assemble a team of people that could do this.

Lex Fridman 49:55 Yeah. How to simulate contact period. So, foot ground contact, but sort of for manipulation, um, because don't you want to model all kinds of surfaces?

Robert Playter 50:06 Yeah. So, it will be even more complex with manipulation 'cause there's a lot more going on (laughs) -

Lex Fridman 50:11 Yeah.

Robert Playter 50:11 ... you know? And you need to capture, I don't know, things slipping and moving, you know, in y- in your, in your hand. Um, y- it's a level of complexity that I think goes above foot ground-

Lex Fridman 50:25 Mm-hmm.

Robert Playter 50:25 ... uh, contact w- when you really start doing dextrous manipulation. So, there's challenges ahead still.

Lex Fridman 50:31 So, how far are we away from me being able to walk with Atlas in the sand along the beach-

Robert Playter 50:37 (laughs)

Lex Fridman 50:37 ... and us both drinking a beer?

Robert Playter 50:40 (laughs) Well-

Lex Fridman 50:42 S- uh, out of a can, out of a can.

Robert Playter 50:43 Maybe Atlas could spill his beer 'cause he's got nowhere to put it. (laughs) Uh, Atlas could walk on the sand. Uh, that's-

Lex Fridman 50:50 So, can it?

Robert Playter 50:51 Yeah. Yeah. Uh, yeah, I mean, you know, have we really had him out on the beach? You know, we take them outside often, you know, rocks, hills, that sort of thing, even just around our lab in Waltham.

Lex Fridman 51:02 Yeah.

Robert Playter 51:02 We probably haven't been on the sand, but I'm, I-

Lex Fridman 51:04 So, soft surfaces are okay?

Robert Playter 51:05 I don't doubt that we could deal with it. Uh, we, we, w- we might have to spend a little bit of time to sort of make that work, but... We did take, uh,

Robert Playter 51:13 we, we had, uh, had to take BigDog to Thailand years ago.

Lex Fridman 51:19 Mm-hmm.

Robert Playter 51:20 And, uh, we did this great video of the robot walking in the sand, walking into the ocean-

Lex Fridman 51:27 Mm-hmm.

Robert Playter 51:27 ... up to, I don't know, its belly or something like that, and then turning around and walking out-

Lex Fridman 51:32 Yes.

Robert Playter 51:32 ... all while playing some cool beach music.

Lex Fridman 51:34 Yeah.

Robert Playter 51:35 Great show, but then, you know, we didn't really clean the robot off, and the salt water was really hard on it. So, uh, we, you know, we put it in a box, shipped it back. By the time it, it came back, we had some problems (laughs) with corrosion.

Lex Fridman 51:45 So, it's a salt, it's a salt water. It's not like-

Robert Playter 51:48 It's salt stuff. (laughs)

Lex Fridman 51:49 It's not like sand getting into the components or something like this.

Robert Playter 51:52 Yeah. Yeah.

Lex Fridman 51:52 But I'm sure if, if this is a big priority, you could make it, like-

Robert Playter 51:55 Right.

Lex Fridman 51:55 ... waterproof or something.

Robert Playter 51:56 Right, right. That just wasn't our, our goal at the time.

Lex Fridman 51:59 Well, it's a personal goal of mine to, uh-

Robert Playter 52:01 (laughs)

Lex Fridman 52:02 ... walk along, walk along the beach. But it's, it's a human problem, too. You get sand everywhere. It's, it's just a giant mess.

Robert Playter 52:07 (laughs)

Lex Fridman 52:08 Uh, so soft surfaces are okay. So, I mean, can we just, uh, linger on the, the robotics challenge? There was, um, there's a pile of, uh, like rubble they had to walk over. Is that, um, how difficult is that task?

Robert Playter 52:24 In the early days of developing BigDog, the loose rock was the epitome of the hard walking surface because you step down, and then the rock, and you had these little point feet on the robot, and the rock can roll.

Lex Fridman 52:36 Mm-hmm.

Robert Playter 52:37 And, and you have to deal with that last minute, you know, change in your foot placement.

Lex Fridman 52:43 Yeah, so you, you step on a thing, and that thing responds to you stepping on it.

Robert Playter 52:47 Yeah. And, and it moves where your point of support is.

Lex Fridman 52:50 Okay.

Robert Playter 52:50 And so, it's really that, that became kind of the essence of the test. And so, that was the beginning of us starting to build rock piles in our parking lots, and-

Lex Fridman 53:02 (laughs)

Robert Playter 53:02 ... and we would actually build boxes full of rocks and bring them into the lab-

Lex Fridman 53:05 Yeah.

Robert Playter 53:06 ... and then we would have the robots walking across these boxes of rocks because that became the essential test.

Lex Fridman 53:12 So, you mentioned BigDog. Can you, can we maybe take a stroll through the history of Boston Dynamics? Uh, so what and who is BigDog? By the way, is who-

Robert Playter 53:23 (laughs)

Lex Fridman 53:24 Do you try not to anthropomorphize the robots? Do you try not to, do you try to remember that they're... This is like the division I have 'cause I, for me, it's impossible. For me, there's a magic to the, to the being that is a robot. It is not human, but it is, it, the, the same magic, uh, that a living being has when it moves about the world i- is there in the robot. So, um, I don't know what question I'm asking, but, uh, should I say what or who, I guess? Who is BigDog? What is BigDog?

Robert Playter 53:57 (laughs) Well, I'll say, uh, to address the meta question-

Lex Fridman 54:00 (laughs)

Robert Playter 54:00 ... we don't try to draw hard lines around it being an it or a him or a her. Um, it's okay, right? People... I think part of the magic of these kinds of machines is by nature of their organic movement of the, of their dynamics, we tend to want to identify with them. We, we tend to look at them and, and sort of attribute maybe feeling to that because we've only seen things that move like this that were alive. And so, um, this is an opportunity. It means that you could have, uh, feelings for a machine. And, you know, people have feelings for their cars, you know, they get a- attracted to them, attached to them. So, that's inherently...... could be a good thing, as long as we manage what that interaction is. So, we don't put strong boundaries around this and ultimately think it's a benefit, but it's also can be, um, a bit of a curse because I think people look at these machines, and they attribute a level of intelligence that the machines don't have. Why? Because again, they've seen things move like this that were living beings-

Lex Fridman 55:13 Mm-hmm.

Robert Playter 55:14 ... which are intelligent, and so they wanna attribute intelligence to the robots that isn't appropriate yet, even though they move like an intelligent being.

Lex Fridman 55:22 But you try to acknowledge that the anthropomorphization is there and try to, first of all, acknowledge that it's there, uh-

Robert Playter 55:31 And have a little fun with it.

Lex Fridman 55:32 ... and have a little fun.

Robert Playter 55:33 You know, our, our more, most recent video, it's just kind of fun, you know, to, to, to look at the robot. We started off the, the video with Atlas, um, kinda looking around for where the bag of tools was 'cause the guy up on the scaffolding says, "Send me some tools."

Lex Fridman 55:50 Mm-hmm.

Robert Playter 55:51 And Atlas has to kinda look around-

Lex Fridman 55:52 Yeah.

Robert Playter 55:53 ... and see where they are, and there's a little personality there that is fun. It's entertaining. It makes our jobs interesting, and I think, in the long run, can enhance interaction between humans and robots that, in a way that isn't available to machines that don't move that way.

Lex Fridman 56:09 This is something to me personally is very interesting. I've been, um, I happen to have a lot of legged robots.

Robert Playter 56:17 (laughs)

Lex Fridman 56:18 (laughs) I hope, uh, to have a lot of Spots in my possession. Um, I'm interested in celebrating robotics and celebrating companies, and I also don't want to, companies that are doing incredible stuff like Boston Dynamics, and I, there's, um... You know, I'm a little crazy. And you say you don't want to, you want to align, you want to help the company 'cause I, I ultimately want, uh, a company like Boston Dynamics to succeed, and part of that we'll talk about. You know, success kinda requires making money. And so, the kind of stuff I'm particularly interested in may not be the thing that makes money in the short term. I can make an argument that it will in the long term, but the kind of stuff I've been playing with is a robust way of, uh, having the quadrupeds, the, the robot dogs communicate emotion with their body movement.

Robert Playter 57:08 Hmm.

Lex Fridman 57:08 The same kind of stuff you do with a, with a dog.

Robert Playter 57:10 Yeah.

Lex Fridman 57:10 But not, not hardcoded, uh, but in a robust way.

Robert Playter 57:14 Mm-hmm.

Lex Fridman 57:15 And be able to communicate excitement or fear-

Robert Playter 57:17 Mm-hmm.

Lex Fridman 57:18 ... uh, boredom, all this kinds of stuff. Uh, and I think as a base layer of function of behavior to add on top of a robot, I think that's a really powerful way, uh, to make the robot more usable for humans for whatever application.

Robert Playter 57:33 I think it's gonna be really important, and, um, it's a thing we're, we're beginning to pay attention to. Um, we really want to start... A differentiator for the company has always been we really want the robot to work. We want it to be useful. Uh, making it work at first meant the, the legged locomotion really works. It can really get around, and it doesn't fall down. And, um, but beyond that, it n- now it needs to be a useful tool, and our customers are, for example, factory owners, people who are running a, a process manufacturing facility. And the robot needs to be able to get through this complex facility in a reliable way, you know, taking, taking measurements. We need for people who are operating those robots to understand what the robots are doing. If the robot gets into needs help or, or, you know, is in trouble or something, it needs to be able to communicate. And a physical indication of some sort, uh, to, so that a person looks at the robot and goes, "Oh, I know what that's, that robot's doing. That robot's going to go take measurements of my, uh, vacuum pump with its thermal camera." You know, you wanna be able to indicate that, and, or even just the robot's, um, about to turn, you know, in front of you, and maybe indicate (laughs) that it's going to turn, and so you sort of see and can anticipate its motions. So these, this kind of communication is going to become more and more important. It wasn't sort of our starting point, um, y- you know, but now the, the robots are really out in the world, and, you know, we have about 1,000 of 'em out with, with customers right now. This layer of physical indication I think is gonna become more and more important.

Lex Fridman 59:21 We'll talk about where it goes 'cause there's a lot of interesting possibilities, but if you can return back to the origins of Boston Dy- Dynamics with sort of the more research, the R&D side, before we talk about how to build robots this scale, so BigDog.

Robert Playter 59:37 So, um-

Lex Fridman 59:38 Who's BigDog?

Robert Playter 59:39 So, the company started in 1992, and in, um, probably, uh, uh, 2003, I believe is when we, uh, took a contract from Dar- so basically 10 years, 11 years. Um, we weren't doing robotics. We did a little bit of robotics with Sony. Uh, they had, uh, AIBO. They had the, their AIBO robot. We were developing some software for that that kinda got us a little bit involved with robotics again. Then there was this opportunity to do a DARPA contract where they wanted to build a, a robot dog.

Lex Fridman 1:00:17 Mm-hmm.

Robert Playter 1:00:18 And, uh, we, we won a contract, uh, to build that, and so that was the genesis of BigDog. And, uh, it was a quadruped, and it was the first time we'd built a robot that had everything onboard that l- you could actually take the robot out into the wild and operate it. So, it had an onboard power plant. It had onboard computers. It had h- uh, hydraulic actuators that needed to be cooled, so we had cooling systems built in, everything integrated into the robot. And, uh, that was a pretty rough start, right? It was just, it was-... 10 years that we were not a robotics company. We were a simulation company, and then we had to build a robot in about a year. So, that was a, a little bit of a rough transition.

Lex Fridman 1:00:58 (laughs)

Robert Playter 1:00:59 Um ... (laughs)

Lex Fridman 1:01:00 I mean, well, c- can you just comment on the, the roughness of that transition? 'Cause, uh, BigDog, I mean, this is this big, uh, quadruped, four legs, robot.

Robert Playter 1:01:11 We, we built a few different versions of them. But the first one, the ear- the very earliest ones, you know, didn't work very well. (laughs) And we would take them out, and it was hard to get, you know, a, you know, a go-kart engine driving a hydraulic pump.

Lex Fridman 1:01:25 Oh, is that what it was?

Robert Playter 1:01:26 (laughs)

Lex Fridman 1:01:26 I was like-

Robert Playter 1:01:27 And, and, you know, uh, having that all work, uh, while trying to get, you know, uh, the robot to stabilize itself, and so...

Lex Fridman 1:01:34 So, what was the power plant? What was the engine? It seemed like, uh, my vague recollection, it f- (laughs) I don't know. It, it felt very loud and aggressive, and, uh, kind of thrown together is what it kind of...

Robert Playter 1:01:48 Oh, it absolutely was, right? We, we weren't trying to design the best robot hardware at the time. And, um, we wanted to buy an off-the-shelf engine. And so many of the early versions of BigDog had m- literally go-kart engines-

Lex Fridman 1:02:03 Mm-hmm.

Robert Playter 1:02:03 ... or something like that. Usually-

Lex Fridman 1:02:04 Are those gas powered?

Robert Playter 1:02:05 Like a, yeah, gas-powered, two-stroke engine.

Lex Fridman 1:02:08 (laughs)

Robert Playter 1:02:08 And the reason why it was two-stroke is two-stroke engines are lighter weight, but they're also ... and we, we didn't, generally didn't put mufflers on them 'cause we're trying to save the weight. We didn't care about the noise. And so (laughs) these things were horribly loud. Um, but we're trying to manage weight because managing weight in a legged robot is always important because it has to carry everything.

Lex Fridman 1:02:28 That said, that thing was big.

Robert Playter 1:02:30 Well-

Lex Fridman 1:02:30 From what I've seen the videos of.

Robert Playter 1:02:31 Yeah, I mean, the, the, the early versions, you know, stood about, I don't know, belly high, chest high. Um, you know, they probably weighed maybe a couple of hundred pounds. But, you know, it ... over the course of probably five years, uh, we were able to get that robot, um, to really manage a remarkable level of rough terrain. So, you know, we started out with just walking on the flat, and then we started walking on rocks, and then inclines, and then mud, and then slippery mud. And, you know, by the end of that program, we were convinced that l- legged locomotion in a robot could actually work 'cause, you know, going into it, we didn't, we didn't know that. We had built quadrupeds at, at MIT, but they were ... they used a giant hydraulic pump, you know, in the lab. They used a giant computer that was in the lab. They were always tethered to the lab. This was the first time something that was sort of self-contained, you know, um, walked around in the world, um, and, and balanced. But ... and, and the purpose was to prove to ourself that the legged locomotion could really work. And so, um, BigDog really cut that open for us. And, and it was the beginning of what became a whole series of robots. So once, once we showed to DARPA that you could make a legged robot that could work, there was a period at DARPA where robotics got really hot, and there was lots of different programs. And, um, you know, we were able to build other robots. We built other quadrupeds to hand ... like LS3, designed to carry heavy loads. We built Cheetah, which was designed to explore what are the limits to how fast you can run. You know, we, we began to build sort of a portfolio of machines and software that let us build not just one robot, but a whole family of robots.

Lex Fridman 1:04:21 To push the limits in all kinds of directions and-

Robert Playter 1:04:23 Yeah, and to discover those principles. You know, you asked earlier about the art and science of legged locomotion. We i- were able to develop principles of legged locomotion so that we knew how to build a small legged robot or a big one. So, the leg, leg length, you know, was now a parameter that we could play with. Payload was a parameter we could play with. So, we built the LS3, which was an 800-pound robot designed to carry a 400-pound payload. And we, we learned the design rules, basically developed the design rules. How do you scale different robot systems to, you know, their terrain, to their walking speed, to their payload?

Lex Fridman 1:05:02 So, when, uh, was Spot born?

Robert Playter 1:05:06 Around 2000 and, uh, 12 or so. So, again, almost 10 years into sort of a run with DARPA where we built a bunch of different quadrupeds. We had a sort of a different thread where we started building humanoids. Um, we, we, we saw that probably an end was coming where the government was gonna kind of back off from a lot of robotics investment. And, uh, w- in order to maintain progress, we just deduced that, "Well, we probably need to sell ourselves to somebody who wants to continue to invest in this, this area," and that was Google. And so, um, uh, at Google, we would meet regularly with Larry Page. And Larry just started asking us, you know, "Well, what's your product gonna be?"

Lex Fridman 1:05:55 Mm-hmm.

Robert Playter 1:05:56 And, you know, the logical thing, the thing that we had the most history with-

Lex Fridman 1:06:01 Mm-hmm.

Robert Playter 1:06:01 ... that we wanted to continue developing was a quadruped, but we knew it needed to be smaller. We knew it couldn't have a gas engine. We th- we thought it probably couldn't be hydraulically actuated. So, that began the process of exploring if we could migrate to a smaller, electrically actuated, um, robot, and that was really the genesis of Spot.

Lex Fridman 1:06:24 So, not a gas engine, and the actuators are electric.

Robert Playter 1:06:28 Yes.

Lex Fridman 1:06:28 So, can you maybe comment on what it's like, um, at Google with working with Larry Page, having those meetings and thinking of what are, will a robot look like that could be built at scale? What ... like, starting to think about a product.

Robert Playter 1:06:45 Larry always liked the, the toothbrush test. He wanted products that you used every day.Um, what they really wanted was, you know, a consumer-level product, something that would work in your house. We didn't think that was the right next thing to do-

Lex Fridman 1:07:05 Mm-hmm.

Robert Playter 1:07:06 ... because to be a consumer-level product, cost is gonna be very important. It probably needed to cost a few thousand dollars, and we were, we were building these machines that cost hundreds of thousands of dollars, maybe a million dollars to build. Of course, we were only building, like, two, but-

Lex Fridman 1:07:22 Mm-hmm.

Robert Playter 1:07:23 ... but we didn't see how to get all the way to this consumer-level product-

Lex Fridman 1:07:27 In a short amount of time.

Robert Playter 1:07:27 ... in a short amount of time. And he suggested that we, we make the robots really inexpensive. And part of our philosophy has always been, build the best hardware you can. Make, make the machine operate well so that you're trying to solve, you know, discover the, the hard problem that you don't know about. Don't, don't make it harder by, by building a crappy machine, basically.

Lex Fridman 1:07:54 Mm-hmm.

Robert Playter 1:07:55 Build the best machine you can. There's plenty of hard problems to solve that are gonna have to do with, you know, under-actuated systems and balance.

Lex Fridman 1:08:02 Mm-hmm.

Robert Playter 1:08:03 And so we wanted to build these high-quality machines still, and we thought that was important for us to continue learning about really what was im- the important parts of, that make robots work. Um, and so there was a little bit of a phil- philosophical difference there that we... And, and so ultimately, that's why we're building robots for the industrial sector now, because the industry can afford a more expensive machine because, you know, their productivity depends on keeping their factory going. And so if Spot costs, you know, um, $100,000 or more, that's not such a big expense to them, whereas at the consumer level, no one's gonna buy a robot like that. And I think we might eventually get to a consumer-level product that will be that cheap. But I think the path to getting there needs to go through these really nice machines so that we can then learn how to simplify.

Lex Fridman 1:08:57 So, what can you say to the almost engineering challenge of bringing down cost of a robot? So that presumably when you try to build a robot at scale, that also comes into play when you're trying to make money on a robot, even in, in the industrial setting. But how interesting, how challenging, uh, of, of a thing is that, sp- in particular probably new to an R&D company?

Robert Playter 1:09:25 (laughs) Yeah, I'm glad you brought that last part up. The transition from an R&D company to a commercial company, that's the thing you worry about. You know, 'cause you've got these engineers who love hard problems, who want to figure out how to make robots work, and you don't know if you have engineers that want to work on the quality and reliability and cost that is ultimately required. Um, and indeed, you know, we have brought on a lot of new people who are inspired by those problems. But, but the big takeaway lesson for me is, we have good people. We have engineers who want to solve problems, and, and the quality and cost and manufacturability is just another kind of problem. And because they're so invested in what we're doing, they're interested in and will go work on, on those problems as well. And so I think we're managing that transition very well. In fact, I'm really pleased that... I mean, uh, it's a huge undertaking by the way, right? So, you know, even having... To get reliability to where it needs to be, we have to have fleets of robots that we're just operating 24/7 in our offices to go find those rare failures and, and eliminate them. It's a, just a totally different kind of activity than the research activity where you get it to work, you know, the one robot you have, uh, to work in a repeatable way, (laughs) you know, at the, at the high-stakes demo. It's just very different. Um, but I think we're making remarkable progress, I guess.

Lex Fridman 1:10:54 So one of the cool things, I got a chance to, uh, visit Boston Dynamics, and I mean, one, one of the things that's really cool is to see a large number of robots moving about, because I think one of the things you notice in the research, uh, environment is that at MIT, for example, I don't think anyone ever has a working robot for a prolonged period of time.

Robert Playter 1:11:18 (laughs) Exactly.

Lex Fridman 1:11:19 So, like, most robots are just sitting there in a sad state of despair, waiting to be born, brought to life for a brief moment of time. But just to have... I just, I just remember there's like a, there's a Spot robot just, uh, had, like, a cowboy hat on. It was just walking randomly for whatever reason. I don't even know. But there's a kind of a sense of sentience to it because it doesn't seem like anybody was supervising it. (laughs)

Robert Playter 1:11:45 Well-

Lex Fridman 1:11:45 It was just doing its thing.

Robert Playter 1:11:46 ... I'm gonna stop way short of the sentience.

Lex Fridman 1:11:48 Sure.

Robert Playter 1:11:48 Um, it is the case that if you come to our office, you know, today and walk around the hallways, um, you're gonna see a dozen robots just kind of walking around-

Lex Fridman 1:11:57 Yes.

Robert Playter 1:11:58 ... all the time, and that's really a reliability test for us. So, we have these robots programmed to do autonomous missions, get up off their charging dock, walk around the building, collect data at a few different places, and go sit back down. And we want that to be a very reliable process 'cause that's what somebody who's running a, uh, a brewery, a factory, that's what they need the robot to do. And so we, we have to, we have to dogfood our own robot. We have to test it in that way. And, um, so on a weekly basis, we have robots that are accruing something like, um, 1,500 or maybe 2,000 kilometers of walking, and, uh, you know, over 1,000 hours of operation every week.And that's something that almost, I don't think anybody else in the world can do 'cause A, you have to have a fleet of robots to just accrue that much information. (laughs)

Lex Fridman 1:12:51 Mm-hmm.

Robert Playter 1:12:51 You have to be willing to dedicate it to, to that test. And, uh, so that's, but that's essential.

Lex Fridman 1:12:58 That's how you get the reliability.

Robert Playter 1:12:59 That's how you get it.

Lex Fridman 1:13:00 What about some of the cost-cutting from the, from the manufacture side? What, what have you learned from the manufacture side of the transition from R&D to ?

Robert Playter 1:13:08 And we're still, we're still learning a lot there. Um, we're learning how to cast parts instill- instead of mill it all out of, you know, billet aluminum.

Lex Fridman 1:13:17 Mm-hmm.

Robert Playter 1:13:17 Um, we're learning how to get plastic molded parts. We're, uh, and we're learning about how to control that process (laughs) so that you can build the same robot twice in a row. There's a lot to learn there, and we're only partway through that process. Um, we've, we've set up a manufacturing facility, uh, in Waltham. It's about a mile from our headquarters, and we're doing final assembly and tests of both Spots and Stretches, you know, at that factory. And, um, and it's hard because, to be honest, we're still iterating on the design of the robot. As we find failures from these reliability tests, we need to go engineer changes, and those changes need to now be propagated to the manufacturing line. And that's a hard process, especially when you wanna move as fast as we do, and that's been, uh, challenging. And it, it makes it, you know, the folks who are working supply chain, who are trying to get the cheapest parts for us, kind of requires that you buy a lot of 'em to make 'em cheap.

Lex Fridman 1:14:17 (laughs)

Robert Playter 1:14:17 And then, we go change the design from underneath them, and they're like, "What are you doing?" And so, you know, getting everybody on the same page here, that it, yep, we still need to move fast, but we also need to try to figure out how to reduce cost. That's one of the challenges of, of this migration we're going through.

Lex Fridman 1:14:32 And over the past few years, challenges to the supply chain. I mean, I i- imagine you've been a part of a bunch of stressful meetings.

Robert Playter 1:14:38 Yeah, things got more expensive and harder to get, and yeah, so it's, it's all been a challenge.

Lex Fridman 1:14:44 Is there still room for simplification?

Robert Playter 1:14:47 Oh, yeah. Much more. And, you know, these are really just the first generation of these machines. Uh, we're already thinking about what the next generation of Spots' gonna look like. Spot was built as a platform, so you could put almost any sensor on it.

Lex Fridman 1:15:00 Mm-hmm.

Robert Playter 1:15:01 You know, we provided data communications, mechanical connections, uh, power connections. Um, but for example, in the applications that we're excited about, where you're, you're monitoring these factories, uh, for their health, there's probably a simpler machine that we could build that's really focused on that use case.

Lex Fridman 1:15:22 Mm-hmm.

Robert Playter 1:15:23 And, uh, that's the difference between the general-purpose machine, or the platform, versus the purpose-built machine. And so, even though, even in the factory, we'd still like the robot to do lots of different tasks, if it's, if we really knew on day one that we're gonna be operating in a factory with these three sensors in it, we would have it all integrated in a package that would be easier, more, less expensive, and more reliable. So, we're contemplating building, you know, a next generation of that machine.

Lex Fridman 1:15:50 So, we should mention that, uh, so Spot, uh, for people who somehow are not familiar, uh, so is a yellow robotic dog and, um, has been featured in many dance videos. Uh, it also has gained an arm. So, what can you say about the arm that Spot has, about the challenges of this design and the manufacture of it?

Robert Playter 1:16:15 We think the future of mobile robots is mobile manipulation. That's where, you know, in the past 10 years, it was getting mobility to work, getting the legged locomotion to work. If you ask, "What's the hard problem in the next 10 years?" It's getting a mobile robot to do useful manipulation for you. And so, we wanted Spot to have an arm, uh, to experiment with those problems. Um, and the arm is, um, almost as complex as the robot itself, you know, and, uh, it's a, it's an attachable payload. Um, it has, you know, several motors, and actuators, and sensors. It has a camera in the end of its hand, so, you know, you can sort of see something. The, and the robot will control, uh, the motion of its hand to go pick it up autonomously. So, in the same way the robot walks and balances, managing its own foot placement to stay balanced, we want manipulation to be mostly autonomous, where the robot, you indicate, "Okay, go grab that bottle." And then, the robot will just go do it using the camera in its hand, and then sort of closing in on that, um, the grasp. But it's, it's a whole 'nother complex robot on top of-

Lex Fridman 1:17:30 Yeah.

Robert Playter 1:17:30 ... a complex legged robot. And so, and of course, we made it, the hand look a little like a head, (laughs) you know, because again, we want it to be sort of identifiable. In the last year, um, a lot of our sales have been people who already have a robot now buying an arm to add to that robot.

Lex Fridman 1:17:50 Oh, interesting. And so, the, the arm is for sale?

Robert Playter 1:17:54 Oh, yeah. Oh, yeah. It's an option.

Lex Fridman 1:17:56 What's the, what's the interface like to work with the arm? Like, uh, is it pretty ... So, are they designed primarily, I could just ask that question in general about robots from Boston Dynamics. Is it designed to be easily and efficiently operated remotely by a human being, or is there also the capability to push towards autonomy?

Robert Playter 1:18:20 We want both. Uh, in the next version of the software that we release, uh, which will be version 3.3, we're gonna offer the ability of, of, if you have a- an autonomous mission for the robot, we're gonna include the option that it can go through a door, which means it's gonna have to have an arm, and it's gonna have to use that arm to open the door. And so, that'll be an autonomous manipulation task that just-... uh, you can program easily, uh, with the robot.

Lex Fridman 1:18:48 Oh, so ƒ-

Robert Playter 1:18:48 Strictly through, you know, we have a tablet interface. And so on the tablet, you know, you sort of see the, the view that Spot sees. You say, "There's the door handle." You know, "The hinges are on the left, and it opens in. The rest is up to you. Take care of it."

Lex Fridman 1:19:02 Oh, wow, so it just takes care of everything?

Robert Playter 1:19:05 Yeah. So, we, we want... and for a task like opening doors-

Lex Fridman 1:19:09 Mm-hmm.

Robert Playter 1:19:10 ... you can automate most of that. And we've automated a few other tasks. Uh, we had a customer who had a, um, a high-powered breaker switch, essentially. It's an electric utility. Ontario Power Generation. And they have to... uh, when they're gonna disconnect, you know, their power supply, right? That could be a gas generator, it could be a nuclear power plant. You know, from the grid, you have to disconnect this breaker switch. Well, as you can imagine, there's, you know, hundreds or thousands of amps and volts (laughs) involved in this breaker switch, and it's a dangerous event 'cause occasionally you'll get what's called an arc flash. Uh, as you dis- do this disconnect, the power, the sparks jump across, and people die doing this. And so, uh, Ontario Power Generation used our Spot and, and the arm through the interface to, to operate this, this disconnect. Um-

Lex Fridman 1:20:03 Mm-hmm. That's right.

Robert Playter 1:20:04 ... in a interactive way.

Lex Fridman 1:20:05 Mm-hmm.

Robert Playter 1:20:06 And they showed it to us. And we were so excited a- about it and said, "You know, I bet we can automate that task." And so we, we got some examples of that breaker switch.

Lex Fridman 1:20:15 Mm-hmm.

Robert Playter 1:20:16 And I believe in the next generation of the software now we're gonna deliver back to Ontario Power Generation, they're gonna be able to just point the robot at that breaker. They'll be a- they'll indicate that's the switch. There's sort of two actions you have to do. You have to flip up this little cover, press a button, then get a ratchet, stick it into a socket, and un- literally unscrew this giant breaker switch. So, there's a bunch of different tasks. And we basically automated them so that the human says, "Okay, there's the switch. Go do that part. That right there is the socket where you're gonna put your tool, and you're gonna open it up." And so you can remotely sort of indicate this on the, the tablet, and then the robot just does everything in between.

Lex Fridman 1:21:00 And it does everything, all the coordinated movement of all the different actuators that includes the body and the arm?

Robert Playter 1:21:04 Ma- yeah, maintains its balance. It, it walks itself, you know, into position, so it could... it's within reach, and the arm is in a position where it can do the whole task. So, it, it manages, uh, the whole body.

Lex Fridman 1:21:17 So, ho- how does one become a big enough customer to request features? 'Cause I personally want a, a robot that gets me a beer.

Robert Playter 1:21:25 (laughs)

Lex Fridman 1:21:26 I mean, that has to be, like, one of the most request- I suppose in the industrial setting, that's, uh, a non-alcoholic beverage-

Robert Playter 1:21:33 (laughs)

Lex Fridman 1:21:33 ... uh, of picking up objects and bringing the objects to you.

Robert Playter 1:21:38 We love working with customers who have challenging problems like this, and, and this one in particular, because we felt like what they were doing, A, it was a safety feature.

Lex Fridman 1:21:47 Mm-hmm.

Robert Playter 1:21:47 B, we saw that the robot could do it 'cause they, they teleoperated it the first time. Probably took them an hour to do it the first time, right? But the robot was clearly capable. And we thought, "Oh, this is a great problem for us to work on to figure out how to automate a manipulation task." And so we took it on not, not because we were gonna make a bunch of money from it and selling the robot back to them, but because it motivated us to go solve what we saw as the next logical step. But many of our customers, in fact, uh, we, we try to... O- our bigger customers, typically the ones who are gonna run a utility, or a factory, or something like that, we take that kind of direction from them. And if they're... especially if they're gonna buy 10, or 20, or 30 robots-

Lex Fridman 1:22:30 Mm-hmm.

Robert Playter 1:22:31 ... and they say, "I really need it to