Launch HN: K-Scale Labs (YC W24) – Open-Source Humanoid Robots

Eh. I think we got a bit nerd-sniped on some things and we ended up trying to build most of our stack ourselves. The control loop is just a pretty simple Rust loop for running ML models. ROS is kind of annoying to work with and the control loop is pretty simple so we just wrote it ourselves.

For two legged - I think it will eventually be quite a bit cheaper, it's mostly a software problem to get them to be stable. RL based control has gotten much, much better. For example, I was talking to Trevor Blackwell a few weeks ago, and he was saying the IMU on the original Anybots robot was over $2k, whereas if you throw a noisy IMU into sim, you can get away with something basically from a cellphone. So yea, personally I'm a big believer in needing to solve the robotics intelligence problem, and once you solve that, humanoids will make the most sense as a form factor.

Feet/ankles on humanoids are really hard mechanically. In many ways the kinematic requirements for the ankle are similar to a wrist, but while the wrist of a robot arm is the least-heavily-loaded, the ankle area can be the most heavily loaded. Humans get away with it by having most of the highly-stressed actuators in the lower leg, not the ankle itself, whereas many humanoid robots end up putting more of the actuators in the ankle assembly itself.

In general, I think the best way to think about the differences between human muscles and robot actuators is that human muscles are simultaneously incredible in terms of strength and power density, and also incredibly fragile. Robot actuators are fairly robust, but comparatively poor. Humans are essentially falling apart at all times, but the repair mechanisms in the body do a good enough job that it doesn't matter (although you probably know someone with a "career-disruptive/ending" sports-related injury that shows these mechanisms have limits). Robotics is a long way away from making actuators that can be fixed online in such a process. Even cable stretching in cable-driven mechanisms remains hard to handle effectively, and that's one of the simplest types of mechanism wear.

So, most humanoids you see that are relatively cost-effective are just "humanoid" in that they look like humans, but there are significant mechanical differences between them and real humans. It's almost always driven by the cost of manufacturing different components. A good example is the lead screw you see in the knee and ankle on Optimus - while it is more human-centric to have tendon-like actuation, it drives the price up quite a bit. Put differently, humans have a lot of specialization which is not great if you want to mass manufacture something.

For walking, the most important thing is that the robot can be simulated well, so in our case, we tried to model the foot contact with the ground in simulation quite accurately. In fact, we found that force sensors in the foot probably help but they're not necessary in simulation, and we wanted to shave off anything that wasn't necessary. I am not sure how close we will get to human mobility - it's definitely a learning process - but you can get much further in simulation than you'd expect.

I have a pretty bad mental model of how most VCs think, but I think good VCs will fund smart people who demonstrate extreme conviction, regardless of how they initially size the market. The opportunity cost for me doing K-Scale is making quite a bit of money at Tesla or Meta, so assuming I am not acting irrationally, either I have extreme conviction or I am a masochist. In my experience, VCs are pretty bad at telling the difference.

Yes, we are aware of this customer segment. I don't think they have thoroughly considered the implications of what high torque actuators can do to the human body

We're exploring some options - maybe a rental option in the future. Would like to make it accessible to everyone

We do! https://bounties.kscale.dev/

We haven't updated it much but it's a good starter point

To be honest, I don't know that there are many applications today which humanoids are the best form factor to solve. I would view it more as a form factor that is likely to get much more useful over the next few years, and having the hardware in your home lets you try out new techniques as they come online.

Personally, I think the first real use cases will mostly be entertainment. Humanoids have a high "coolness" factor. Also, I think there's a long tail of random problems which you don't want to buy a new robot to solve, but which, if you have a robot lying around that isn't perfect but is "good enough", might be possible to solve imperfectly. For example, I just had a newborn baby, and I was thinking it would be nice if I had a static arm that could hold his bottle for me. There's a lot of tail end problems like that in your day to day life. But I think the really interesting capabilities will come once there's very good end-to-end models running on-device.

I spent two years on Tesla's FSD team, and I think from a cash flow perspective for funding R&D this model did make a lot of sense - basically, it takes cash upfront for training models, but there's zero marginal cost for distributing the models once you've developed them.

I think this kind of "promise the future, pay now" model does alienates some people, especially when the tech is not ready today. That's why we're open sourcing everything, to avoid the feeling of overpromising on what is ready today. The core idea is that the people who bought FSD early on were very invested in it's success, and that feedback loop is very important for improving machine learning models at scale. The problem happens when actually delivering on the tech takes a long time, but I think we have a fairly clear technical roadmap to make our robot useful. At least, I think there are a lot more intermediate benchmarks for driving value for a humanoid robot than there are for self-driving cars, so I think people who buy it will have a stronger feeling that it is constantly improving.

Thanks!

We're all pretty cross-stack - there are some hardware people and some software people, but the product is quite integrated. Personally, my time has been mostly focused on the RL stack recently, and after there are more robots in the wild, I suspect my time will transition to working on building this data feedback loop.

I try to answer questions pretty actively on our Discord so happy to chat there about whatever you like