The Robot Report Podcast
00:52:22 · Steve Crowe
[00:52:22] S0: Are there things holding back how quickly you guys could scale these behaviors on Atlas to more tasks or even broader tasks? Like, I think you guys use the new Jetson Thor platform, but is it computing power? Is it training data and diversity of that data? Like are those sort of, I don't mean this in a negative way, but what are those bottlenecks that are holding you guys back from taking this even further more quickly?
00:52:50 · Scott Kuindersma
[00:52:50] S4: Yeah, for sure it's data. That's just because in order to collect, like the highest quality data that you can get is data for the tasks you care about on the robot you care about. And there's just a limited amount of that. So you only have so many humanoids. Humanoids are among the most complex robots that anyone's trying to do this kind of work on. So, I think as you look at other types of robot platforms like table mounted arms and things like that, I think your ability to scale it up to just massive data collect facilities is a little better than it is with humanoids. So, I think that's one thing. But one of the really exciting things in this space is that there's other scaling bets that we can leverage here. So I think if the only answer we believe was true is that you needed on hardware experience in order to train these models and nothing else would work. That would be pretty problematic because I think in order to get to even like GPT, I don't know, two level scale would be a tremendous undertaking, right? And so what people tend to talk about is something like a data pyramid. And so you could think about the tip of that pyramid as being the highest quality data that if you just had more and more of that, things are only gonna get better. And so that's sort of like the teleoperation on Atlas. And then as you go down that pyramid, the scalability proposition gets a lot better, but the embodiment gap maybe gets bigger as well. So maybe if I go down a little bit, I can talk about simulation data, right? So if I go down somewhere in the pyramid, having simulations of my robot doing manipulation tasks could be used as co training data. So this is data that we throw into the mix with the real robot data. We've actually found in our experience that, and other groups have too, that incorporating simulation data as part of the training mixture actually improves performance on the physical robot, which is really exciting. One of the other opportunities about humanoids specifically is maybe there's more of an opportunity to use human demonstration data that isn't teleop. So if I have recordings of people, people wearing instrumentation, egocentric videos of people doing manipulation tasks, if their bodies are not too dissimilar from my robot's body, that may be an easier thing to translate into and inform robot behavior.
00:55:17 · Scott Kuindersma, Eugene Demetri
And then you go all the way down to probably the bottom of the pyramids like YouTube. So it's like very unstructured video of people doing things and it's just a harder thing probably to leverage into real concrete improvements in robot behavior. But it'll probably be part of the story. So from our perspective, we're thinking about that whole set of things. And I think, you know, so is many other groups. And I think everyone's sort of taking different bets on the relative proportions that they're expecting to be part of the right answer, and they're investing resources in different parts of that pyramid. So you see people with like giant factories filled with teleoperators, and then some groups with like relatively modest tele operations, but a lot of investment around egocentric data collector simulation or things like this. [00:56:10] S3: And you mentioned before that obviously, Toyota Research Institute as a partner has been helping with that. Can you talk about that ongoing relationship for a moment? Isn't it something that you're expecting to continue?
00:56:10 · Eugene Demetri, Scott Kuindersma
[00:56:10] S3: And you mentioned before that obviously, Toyota Research Institute as a partner has been helping with that. Can you talk about that ongoing relationship for a moment? Isn't it something that you're expecting to continue? [00:56:21] S4: Yeah, for sure. Yeah, we're continuing to work together. I think us working with them has been just so much fun. They're an awesome team. I think we are very similar in terms of our outlook and what we think is important. The structure of how we've been working together is, we have a special space in our building that's kind of just like a research lab, and there's research scientists from TRI who are mixed in with our own research scientists in there, and they all kind of just work as one big team. We'll get some Atlas robots in there. So it's been really great, and I'm looking forward to continuing to work with [00:56:56] S3: And are you also working with the AI robotics institute as well?
00:56:56 · Eugene Demetri, Scott Kuindersma
[00:56:56] S3: And are you also working with the AI robotics institute as well? [00:57:00] S4: Yeah, for sure. So, yeah, that's kind of our sister organization, if you'd like to call it that. Yeah, so one of... There's many ways in which we work together. I think one of the most visible ones is our collaboration on reinforcement learning. So, that's another project that actually exists in the same physical space. So, we have people from the Robotics and AI Institute who are hanging out with us in the lab, and then we have a bunch of our own research scientists who are working with them. And so the results that we showed earlier in the year with the robot doing cartwheels and break dancing and things like that was actually a result from that collaboration. So the way I think about TRI collaboration is more focused on the LBM style manipulation research. And then the other project that we have that's a similar mechanically as that is the... With Ray is the reinforcement learning project. [00:57:54] S3: I may be able to answer this, but I know that Hyundai has been talking about testing Atlas as well. What's going on with that right now? [00:58:04] S4: Yeah. So that's something we're super excited about. Obviously, one of the big... You know, Hyundai is likely to be our our biggest, most important first customer as as they announced. And so we're really excited about Atlas doing work in their facilities. So we're planning on taking Atlas to some Hyundai facilities this year, and I'm sure we'll tell everyone about it when we're ready.
00:58:04 · Scott Kuindersma, Steve Crowe
[00:58:04] S4: Yeah. So that's something we're super excited about. Obviously, one of the big... You know, Hyundai is likely to be our our biggest, most important first customer as as they announced. And so we're really excited about Atlas doing work in their facilities. So we're planning on taking Atlas to some Hyundai facilities this year, and I'm sure we'll tell everyone about it when we're ready. [00:58:27] S0: That's awesome. Can't wait for that. They must be... I know they're... They just made a couple massive announcements about manufacturing back here in The States and how much robotics will be a part of that. They sound like a great partner. Just to wrap things up, I know you've obviously seen the evolution of Atlas in your time at Boston Dynamics. You've seen the evolution of these new training techniques. What's next for Atlas? What can we see besides the Hyundai trials? And how excited are all of you guys about where this is all headed? [00:58:56] S4: Yeah, I mean, we're very excited. I think you're gonna see more and more AI centric results from us. I think we're really committed to trying to break through the ceiling for humanoid manipulation. So I think you should look forward to more exciting things, both on the manipulation front as well as on the very on brand dynamic behavior front.