Hacker News Reader

Comment by ActionHank | Hacker News Reader

Every single one of these robot plays are going to try pivot to having someone do you laundry remotely for a dollar a day.

If they ever get the tech down (I doubt it), they will then try move to broader labour replacement.

This is why all the robots need to be humanoid.

View on HN

Curious why you feel they won’t get the tech down? I think these products are all data plays right now.

Manipulating soft bodies is exceptionally difficult to do with robotics.

moving a soft object from a to b is doable, folding/separating/sorting at any kind of speed is very much an area of active research.

Totally agree but the idea is this gives you a teleoperation environment that is truly on policy and not some artificial lab. The idea is that these robots, like those Amazon stores, are predominantly just controlled by actual humans.

Yeah but if it’s a person seeing inside my home I would rather it be the same person over and over and without cameras.

Teleoperated robots are a better fit for businesses and public spaces

Hey I totally agree I do not want a teleoperator looking into my house, it’s just so deliciously tempting to get in home on policy data. Not sure the reason why they are super interested in home environments vs business or public spaces.

How about sensory feedback loops?

There are at least two companies out there that can fold some laundry in a controlled environment already. At this point it's just a matter of categorizing fabrics and shapes and expanding that knowledge. Two years at the outside until it can fold 80% of your laundry.

Forgive me if I am somewhat sceptical.

folding a flat, well presented tshirt, that is a known quantity, and never inside out, is fairly "easy"

in an unknown environment, with unknown clothes, that's still active area of research. Its still not that practical to navigate in a house yet.

Combined with:

> To get it right, we designed and built our own actuators, remote actuation system, and safety systems.

I feel that there might be a degree of ... overconfidence here.

I'm curious what they felt was lacking in current commercial/industrial robotics safety systems.

But the $1 per day is soooo cheap. You can't even run a local LLM for that price.

Latency, unless we get instantaneous comms working there will always be lag.

Annoying when you play games, likely expensive when the robot keeps breaking things.

This is basically a solved problem. Humans have latency, more than your Internet connection. We predict what's going to happen based on the shapes we see. Computer vision and LBMs do the same thing.

Reply on news.ycombinator.com