The problem is that these are under ideal conditions. And I don’t see an application for a running robot that can operate on under ideal conditions. Show me this thing doing the same thing under adverse conditions, and actually having an application (delivery?) and I’ll ne impressed.
I acknowledge that this is a technical feat, and not an easy one. But show me why it matters. Why this is better than a wheeled robot moving at the same speeds.
Running merely illustrates that the system can react with very little latency, it’s obvious that this will be applicable in any applications where the robot needs to quickly adapt to the environment, such as say factory work.
In à set a planned condition you don’t have the impact of random events. A perfectly fine program can broke easily because input data don’t match expected input. I join the other guy sating that not a huge feat, that a feat but on the same scale of solid state battery or the folding cloth robot. Pretty but not real life usable yet so not so usefull.
I get that you like the final idea. I am myself wetting myself about the LG robot which was folding t-shirts at CES2026 but I fold on my bed and I am convinced he can’t handle well a tilted surface which is not as hard as a table.
You absolutely do have the impact of random events when you’re doing anything in the physical world. You have wind, uneven ground, variations in weight distribution, and so on. That’s what makes this sort of stuff so difficult in practice. All the tiny little errors quickly add up, so you can’t just match expected input. You have to have a dynamic system that can adjust on the fly to the sensory data. Dealing with stuff like an uneven bed or a tilted surface is a completely separate problem of having a good enough world model internally.
I may be not knoweldable enough in AI but doesn’t a bed on which you have to fold thing part of the world model ?
If I understand that a break through in the mécanisme which do micro ajustement of the derive.
Yes, the bed and the environment in general is part of the world model. What I mean is that’s part of object identification and recognition of what objects to use for what task, etc. It’s a separate concern from dexterity. Think of it this way. If you’re thirsty, and you pick up a cup. You’re consciously thinking about moving your hand to grab the cup and bring it to your mouth. That’s what the world model is concerned with. You’re not aware of every individual muscle movement and all the micro adjustments that need to happen in order for the task to be completed. And that’s what the running illustrates. It’s the dexterity of the system in dealing with feedback from the world and making these adjustments in response.
The problem is that these are under ideal conditions. And I don’t see an application for a running robot that can operate on under ideal conditions. Show me this thing doing the same thing under adverse conditions, and actually having an application (delivery?) and I’ll ne impressed.
I acknowledge that this is a technical feat, and not an easy one. But show me why it matters. Why this is better than a wheeled robot moving at the same speeds.
Running merely illustrates that the system can react with very little latency, it’s obvious that this will be applicable in any applications where the robot needs to quickly adapt to the environment, such as say factory work.
I disagree, but I’m just a mechanical engineer.
And what specifically is it that you disagree with, but I’m just a software engineer.
In à set a planned condition you don’t have the impact of random events. A perfectly fine program can broke easily because input data don’t match expected input. I join the other guy sating that not a huge feat, that a feat but on the same scale of solid state battery or the folding cloth robot. Pretty but not real life usable yet so not so usefull. I get that you like the final idea. I am myself wetting myself about the LG robot which was folding t-shirts at CES2026 but I fold on my bed and I am convinced he can’t handle well a tilted surface which is not as hard as a table.
You absolutely do have the impact of random events when you’re doing anything in the physical world. You have wind, uneven ground, variations in weight distribution, and so on. That’s what makes this sort of stuff so difficult in practice. All the tiny little errors quickly add up, so you can’t just match expected input. You have to have a dynamic system that can adjust on the fly to the sensory data. Dealing with stuff like an uneven bed or a tilted surface is a completely separate problem of having a good enough world model internally.
I may be not knoweldable enough in AI but doesn’t a bed on which you have to fold thing part of the world model ? If I understand that a break through in the mécanisme which do micro ajustement of the derive.
Yes, the bed and the environment in general is part of the world model. What I mean is that’s part of object identification and recognition of what objects to use for what task, etc. It’s a separate concern from dexterity. Think of it this way. If you’re thirsty, and you pick up a cup. You’re consciously thinking about moving your hand to grab the cup and bring it to your mouth. That’s what the world model is concerned with. You’re not aware of every individual muscle movement and all the micro adjustments that need to happen in order for the task to be completed. And that’s what the running illustrates. It’s the dexterity of the system in dealing with feedback from the world and making these adjustments in response.
I see, thanks for taking to time for explaining.