Podcast: Human-shaped robots - what's the point?

Kal Mos: At the end, it's extremely valuable to have a humanoid form factor. The whole world, as everybody's saying, the whole world is designed for humans, right? We designed this chair for us, we designed this table for us, we designed the transportation for us, we designed the factory pipeline for us. The whole world is designed for humans. So if you can create a machine that acts like human, looks like human, of course it will be very, very valuable.

Now, until you reach that at a reasonable cost, it may not be the starting point, though. You might have more value in the short term for a machine with a wheel, as you saw on the show floor, because that's easier to move around. If you replace the legs or the double paddle, you can actually put in that thing the battery, so you can have a bigger battery. And you don't have to deal with a lot of stability issues because these things keep it more stable. So there's a lot of concerns about the feet, for example. So that's the first thing that you might want to think about compromising on in the short term and say, okay, maybe we start with maybe something that moves on wheel as a starting point. And then when I solve the problem of battery, when I solve the problem of balance and all of this stuff, maybe I go to the humanoid.

But my strong belief is that the humanoid, as you can see it online today on all this YouTube video, is the right form factor. It is the right target. And eventually when we create that at higher cost with good battery lifetime and all of these things, it will be the right goal.

Dennis Scimeca: What function does the torso serve on a humanoid shape robot?

Kal Mos: So what do you want to replace it with?

Dennis Scimeca: Let's use the example of the Cobot arm that's on the floor right now. It doesn't make a torso. All it needs are the arms. So what's the torso? If you have a humanoid robot, now you've just added a torso, you've added an extra arm, two legs and a head. You just made the robot much more complicated than it used to be. It's true. So why do that if you can use a simpler machine that only has the functionality it requires?

Kal Mos: I think you're absolutely right again on the short term as we get into that. And I go back to the question that this whole world is not designed for these new machines, it's designed for us. So that's why we're trying to get to a machine that looks like us, because it makes it possible for us not to change everything around us.

If you look at this machine that you see now without a torso, it's just arms on a moving wheel, it's perfect for certain function, it will help us in the short term. But now as you start thinking about, does it fit in every area where humans are working on today? Probably not. Does it fit in the transportation system that we have for workers? Maybe not. Does it fit in, like, there are lots of things that are designed around us as human.

So, and as I started, it's really a value versus cost. So for today, I think this is a good machine to start with and it will help us in addressing so many of the use cases. But if we can reach another level where we can create a full humanoid form factor, I think this is the right eventually target for us.

Rehana Begg: On the flip side of that, the other thing that I did not see much of, we saw lots and lots of humanoids, is the quadruped robot, like Spot. I did see one and I talked to someone about what they were being used for, are they still as popular as before, and they are, especially when it comes to safety environments, sniffing out gases that they're used for. The vision quality has been improved that I've been watching year over year. So they're able to do more things using vision technology and artificial intelligence, learning from models and what different types of things that they can sniff out, pardon the pun. So that was pretty interesting. 

Dennis Scimeca: Yeah, but they're all so small. I think that was one of the differentiators as a company over there. They made one that's 165 centimeters tall, I think, which is roughly human height. It's the only one I've seen. They've all been small. I don't understand that. So we want to create this thing that looks human, but it's tiny. Most humans aren't tiny. So just, I don't know. I don't understand a lot of the logic behind it. It's fun to watch them fall. That's always good.

Rehana Begg: I actually went and asked, the humanoid robotic maker, are you going to be improved? How have you improved over the last time I saw this? You know, this really was just for demonstration. The fact is that everybody is delving into it. They need to understand more about it. They're learning about the technologies and the components that go into these robotics. And right now, I think I certainly feel that I've seen a whole lot more use cases, not necessarily out of the pilot stage yet, but many more places where you could potentially have a humanoid do pick and place. Yeah, still doesn't account for the form and function.

Dennis Scimeca: Yeah, that's the classic example. Someone showed me robots to pick and place. Said, but yeah, you could do that with a six-axis roam and a camera. Why do I need this thing? No answer. I never get any answers from anyway.

About the Podcast
Great Question: A Manufacturing Podcast offers news and information for the people who make, store and move things and those who manage and maintain the facilities where that work gets done. Manufacturers from chemical producers to automakers to machine shops can listen for critical insights into the technologies, economic conditions and best practices that can influence how to best run facilities to reach operational excellence.

Listen to another episode and subscribe on your favorite podcast app