RoboPapers
00:23:00 · Michael Cho, Xiawei Liu
[00:23:00] S0: And I guess, you know, you mentioned about the... A lot of it is is... Hinges on kind of the skills of of the tally operator. How, like, how's the actual experience for you guys who... Yeah. I I imagine each of you put in hours of, like, teleoperating. Right? How is it like like, it's... Is it really the case that... Or or rather how many hours of training or or practice does it... Before you guys feel that you are actually maximizing, like, pushing to the limit of what your, you know, whole body teleopset can do? [00:23:32] S4: I would say for teleoperators, there's a few tricks that someone could, you know, get really hands on really fast once they know it. Basically, maybe I would say a good two hour... Under two hours of training would would help, you know, students to get hands on with teleoperating. Because the really hard part with teleoperating is that since the g one's neck couldn't move, students needs to make sure their neck their neck doesn't move. Because if their hands are at the same place, if they move their neck, for g one, their hands would be moving to the other direction that the head is moving. So, yeah, it's it's really tough for students to not move their neck when they're teleoperating because if they wanna grab something, for example, on the right side of the desk, they would intentionally just, like, you know, rotate their neck, and it's really hard for them to adapt to that at first. But after one hour or two hours of training and calibrating, they would be getting used to it. Yeah.
00:23:32 · Xiawei Liu, Michael Cho
[00:23:32] S4: I would say for teleoperators, there's a few tricks that someone could, you know, get really hands on really fast once they know it. Basically, maybe I would say a good two hour... Under two hours of training would would help, you know, students to get hands on with teleoperating. Because the really hard part with teleoperating is that since the g one's neck couldn't move, students needs to make sure their neck their neck doesn't move. Because if their hands are at the same place, if they move their neck, for g one, their hands would be moving to the other direction that the head is moving. So, yeah, it's it's really tough for students to not move their neck when they're teleoperating because if they wanna grab something, for example, on the right side of the desk, they would intentionally just, like, you know, rotate their neck, and it's really hard for them to adapt to that at first. But after one hour or two hours of training and calibrating, they would be getting used to it. Yeah. [00:24:51] S0: How about... How do you guys think about the different additional sensors that you you guys put? Right? I I I noticed you guys also have a third person exocentric view, and then you you put some LiDAR. Could could you talk about, like, that process? How do you think about what, you know, modalities to collect, why they're useful, and things like that?
00:24:51 · Michael Cho, Zhenyu Zhao
[00:24:51] S0: How about... How do you guys think about the different additional sensors that you you guys put? Right? I I I noticed you guys also have a third person exocentric view, and then you you put some LiDAR. Could could you talk about, like, that process? How do you think about what, you know, modalities to collect, why they're useful, and things like that? [00:25:09] S2: So here we... For the LiDAR, it's it's really all baked in into the the G1 system. We were just trying to maximize all the modalities that we can take out of the the G1 and and the H1. So the... So they they both have the the LiDAR, yeah, at their head. [00:25:27] S0: I see. And and I suppose, like, you know, so you guys did so much work to to gather this dataset, and then you you guys then do this, like, actual testing, you know, as you show in this slide here. How has that dataset, like, you know, affected these results in your view? Is there anything surprising that you you didn't expect?
00:25:27 · Michael Cho, Hongyi Jing
[00:25:27] S0: I see. And and I suppose, like, you know, so you guys did so much work to to gather this dataset, and then you you guys then do this, like, actual testing, you know, as you show in this slide here. How has that dataset, like, you know, affected these results in your view? Is there anything surprising that you you didn't expect? [00:25:50] S3: Yeah. So so I think, first, first of all, we we found that all the traditional imitation learning methods are are not working well on on on the majority of our humanoid data. And and we first think that we we we do not collect the the, you know, the correct data. Maybe maybe the data quality is not that high. So so we then try to, like, collect more accurate data, and we then train them with the same imitation learning method, but we we still find out that maybe maybe this imitation learning method just cannot work well on this, you know, humanoid dataset. So then when we turn to train all those VOA... Large VOA models, we found that, wow. This this OpenVLA actually works very well on the humanoid manipulation. And then we're to tune the, you know, the openpi methods, like the pi 0.5 here and the pi 0.5 model here. And we we also find that, well, actually, the robot can can, you know, do the manipulation task very well compared to the imitation learning methods. And and finally, when we try to train these good models here, we find that the NVIDIA GR00T model is the best among all these VLA models. And, yeah, I think that's what we are getting from the list training process.
00:25:50 · Hongyi Jing, Michael Cho
[00:25:50] S3: Yeah. So so I think, first, first of all, we we found that all the traditional imitation learning methods are are not working well on on on the majority of our humanoid data. And and we first think that we we we do not collect the the, you know, the correct data. Maybe maybe the data quality is not that high. So so we then try to, like, collect more accurate data, and we then train them with the same imitation learning method, but we we still find out that maybe maybe this imitation learning method just cannot work well on this, you know, humanoid dataset. So then when we turn to train all those VOA... Large VOA models, we found that, wow. This this OpenVLA actually works very well on the humanoid manipulation. And then we're to tune the, you know, the openpi methods, like the pi 0.5 here and the pi 0.5 model here. And we we also find that, well, actually, the robot can can, you know, do the manipulation task very well compared to the imitation learning methods. And and finally, when we try to train these good models here, we find that the NVIDIA GR00T model is the best among all these VLA models. And, yeah, I think that's what we are getting from the list training process. [00:27:06] S0: I'm curious. Do you have some ablation study where you you do this, like, just straight out the box? Like, can can GR00T or PIE do any of these?
00:27:13 · Hongyi Jing
[00:27:13] S3: Right? Actually, we we we do some ablation study here. And and, you can can you, like, switch it to the paper so we can we can talk about it? Yeah. Yeah. Yeah. So so, yeah, you can see here. So so we do this episode in this ablation experiment, we want to validate that our humanoid everyday data is really useful, can can serve as a useful prior in training those VOA models. So so, actually, when tuning those VOA models, we do this ablation study. The first the first the first baseline is that we directly tune this, like like like, I'll say, like, OpenVLA, direct tune it with new new task, like inserting the rows into the waste. And the second choice is to first tune this OpenVLA model with all the Humanoid Everyday Data, like, 260 tasks, and then tune it with this inserting the results into the waste task. So comparing these two methods can... Like, we can see that when when trained first... Pretrained first with our Humanoid Everyday Prior, we can see that our... The whole performance is actually better than the one without our Humanoid Everyday Prior. So, yeah, you can you can see it here. Like, so the... For the pi 0.5 model, if if... Without the Humanoid Everyday Prior, we can we can never get any success rate in this task. But but with our Humanoid Everyday Prior, we can achieve actually four out of 10 to succeeding this task. And, actually, for obvious VOA models, we all find the the same... You know?