Launch HN: Nori Robotics (YC S26) – A low-cost humanoid robot for development
Hey HN, I’m Antonio from Nori Robotics (https://norirobotics.com). We build a $1,688 bimanual mobile robot in San Francisco for robotics developers and researchers. I started working on Nori while doing robotics research at Columbia. I was teaching robots through human demonstrations, but getting my hands on affordable hardware was difficult. Most labs have one or two expensive robots, which makes it hard to collect large datasets, run long experiments, or test across several robots. So I built my own. After seven iterations the latest Nori has: * 19 degrees of freedom * Two 7+1 DOF arms with a 1.5 kg payload per arm * A 55 kg telescoping lift * A differential wheeled base * Four 720p, 30 fps RGB cameras * 2D lidar * A dual microphone array with full-duplex voice communication * A 432 Wh battery * A Raspberry Pi 5 with 4 GB RAM (SLAM and safeties are run on board, heavier ACT and VLAs must be run from a computer via LAN or a server via WAN) Getting this under $2,000 was the main engineering challenge. Nori has more than 100 moving and structural parts, so costs add up quickly across actuators, bearings, wiring, power delivery, and assembly. Some main choices we made to get the cost low was using high-ratio servos instead of QDD motors, and using a wheel base instead of legs. We assemble each robot in San Francisco and have designed it to be easy to manufacture and repair (we offer 3D files to print repairs). Our open SDK includes teleoperation and demonstration tools: https://github.com/Nori-Robotics/nori-sdk-py We also built a browser-based simulator so you can try it out: https://lab.norirobotics.com/nori/model We’ve shipped our first robot and are building the next batch. Eventually, we want people without robotics experience to teach Nori tasks and share them with other owners. Currently the hardware is already capable of basic cleaning tasks, opening drawers, restocking shelves and pouring beers. Here is a video of Nori doing things: https://youtube.com/shorts/VRfVXHfQvD8 We make money by selling the hardware for $1,688, with optional paid software on top. Parts of hardware are open source. More details are in our hardware paper: https://doi.org/10.48550/arXiv.2605.16537 If you work in robotics, what would you build with a robot at this price? What would you change about the hardware?
Comments
5 preview comments · loading full threadLog in to h4cker, then connect Hacker News to publish comments.
The biggest problem is they are using RC style servos. This is why all the arm motions are jerky and lack precision. What this means: No force feedback on positioning. Jerky motion due to actuator steps. Limited precision. Terrible slow motion control performance. Fast movements will look better. Inability to solve problems with software. Basically think of all the joints like those cheap thermal camera screens that are at 320 x 240 resolution. For rough work or finding some sort of oddity your good, but anything precise you need to spend real money. It's a pretty cool toy, but don't go buying this expecting that much.
Looks cool. Can you be up front about its real capabilities. Are these videos of it eg tidying up real or just staged / cherry picked? If you let it loose in a messy room or stirring something in a kitchen, what are the genuine results you would expect in the wild? We’re all used to seeing things like Atlas back flipping but then can’t do anything in a real environment. Just curious how this fits in and what the success / fail rate is.
Where are you guys in SF? Would love to come see it. Mostly interested in seeing the speed of action (what multiple is the video replay at, etc.) and looking at center-of-mass etc. issues. I have a pretty straightforward use-case in a home. The Dreame we have in the living room is pretty good at avoiding most obstacles, but I usually have an LLM-based agent review the image automatically and either send out the vacuum or suggest moving things that would foul it up. Works great. Obviously substituting with a humanoid capable of reaching ground height things and picking up children's toys etc. would be great. I'd want to bolt it with a mount to the wall when not in use because it needs to be climb-resistant for children during the daytime when I don't need it functional but there is a solid 8 hour 2200->0700 window when no one will be in the living room and speed is not an issue. I was talking to my wife about this device and actually the wheeled form factor and (probably) no control software for balance is a feature for our use-case because when the Unitree G1 or whatever falls over the result is catastrophic. The balance control software attempts pure insanity and will likely kill a child nearby.
1. Does this require internet connectivity? If Nori goes out of business, can we keep using these? 2. Is the app distributed as an APK outside of the Play Store? Does it have any hard requirements on GMS that would prevent users of degoogled devices from using it? 3. Is every function available through the app possible to perform or recreate with the open SDK? Is this robot (now, and in the future) going to be accessible to technically competent folks who don't own a smartphone at all? 4. Is there a privacy policy on your website for the robot itself? Is there any telemetry? What kind? Will it be recording me without me asking it to? Is the software that ships on it totally open source? Cool project! I might be quite interested, depending on how much it respects user freedoms and privacy.
So many moving parts that can break. I am already sad my robot vacuum cleaner broke (it's the moving lidar that broke). I have to send the entire thing including dock back to the manufacturer (very expensive). My preference would be ifixit level guides where you can easily replace parts yourself. Then it would be perfect if the same servo is used everywhere so I can have 3 or 4 stored somewhere.