Launch HN: Nori Robotics (YC S26) – A low-cost humanoid robot for development (norirobotics.com)

Love the design and freedom of this! Curious what the business model of this is to make the company sustainable. Asking that because I was part of the K-scale labs community and sad to see it shut down due to the competition with low-cost chinese robots

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?

Neat vision, but I can only imagine this thing literally fucking everything up. Like, knocking shit off my table, setting my kitchen on fire, bumping into things like an idiot.

Not sure who the market is here apart from SF tech bros with money to burn.

FYI: I backed off the page after about 5 seconds because the scrolling is unusable.

Is there a term for the design style of the plastic panels? I've been seeing it more often, for instance the robot wheeled base looks like an Amazon Delivery Van. Hidden screws, 45degree angles but with soft corner bends.

This is cool! As someone who's built based off of old Roomba platforms and other models, I feel like this looks pretty fun and a good price point. What do you think the next version will target? Will you stay educational and low price point? Targeting some other usecases?

awesome!

Could it put small to medium sized packages in a package room of a condo?

Hopefully upto 24" in any dimension and 5-10 lbs(and ignore the others it can't handle)?

The set up of the small room: three stacked total height 5' rack shelves on both sides and a 2' wide passage in the middle.

Not saying off the shelf but how hard would it be to teach it how to do that?

My question would be how riffable is it? I don't have 10-20k to drop on a robot. I work for co-ops so while I make good money(90k~ pretty good!) its less than I could make else where but the trade off is I have more time to build random stuff. 2k is the price point for me where I might buy a big thing every so many years, teaching this to play board games, making new hands for it, teaching it to clean etc would be fun and at a price point that works for me. I would probably want a suction attachment for the fingers so it could handle cards and other flat tokens but that seems simple enough to build given the 3d models for printing are included.

I could see myself printing all sorts of costumes for it and using it as an ad hoc characters for tabletop games or as a spare bandmate if I ever finish my synth.

"We make money by selling the hardware for $1,688, with optional paid software on top."

The software would have to be quite good and also affordable for me to consider it, as all subscriptions are liabilities. I might one day want my robot to go with me to a show or to do something meaningful. Whenever that happens I will NOT be held hostage by software and will either source my own software or replace parts of the hardware till something works before I keep throwing money at a bad service. I would prefer to just have a good vendor supported software that I don't have to think about but I have trust issues. Privacy would be a huge concern, I can't have this in my garage lab with all my stuff being sent back to your servers for computer vision tasks etc.

Congratulations! That price point is really tempting. I look forward to reading more about this robot.

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.

Quick observations about website. First visited under LibreWolf in a container window. I know privacy/security settings are maxed in this FF clone browser. This is what I could see:

First panel loaded and Nori moved in the demo window. Then I scrolled down to the Everyday Tasks pane. On first visit, this pane took a long time to load. Once it loaded though it did walk through the repeating demo videos. Scrolling down to the "Teach Nori to support you in every way" panel where none of the content loaded. It displayed three black panes with zero content. Continued scrolling down to the "Skills Marketplace" panel and again, nothing in that panel loaded or played. Scrolled down another page to a black panel with no content. The first time I visited, this section did load revealing the eyes widening and gripper extending to take on a new task. Scrolling down more there is a set of attributes in the "Affordable and Capable" panel. Each of the attributes loaded correctly AFAIK. Scrolling to the final panel "Based in the USA" where you can click through to Order, that panel loaded with no apparent issues.

I repeated this using the current incantation of FF and each panel loaded quickly including the "Teach Nori..." and the "Skills Marketplace" so I was able to see what all that was about.

Now to the robot.

First impression is that this is an indoor-only model based on the configuration of the wheels and the lack of weatherization of the joints with wire harness poking out where it can hang on nearby branches, etc. so you need an open environment. Circuit boards in the video shorts show exposed electronics, probably the camera that assists the grabbers in object ID. That probably is not good. The heat-shrink harness on the back that flexes as it extends looks like another way to pull the robot over so I'm thinking this thing won't navigate furniture that has features extending far enough from the unit to enable passing people or robots to become hung.

The video short of the Nori clearing the table is interesting. The specs for the arms indicate a 1.5 kg capacity (3.3#). I looked up the weight of a standard plastic kitchen trash can and they seem to be about 0.45 kg (1#). I see Nori loading light items into the can and that raises a question - Does Nori know when the can is too heavy for it to lift? Does Nori have sensors or will the object detection allow it to track the probable weight of objects in the can so that it doesn't overfill and dump the can contents on the floor? Will Nori understand that some operations require use of both arms? Since Nori is an inside robot, who gets to tote that trash bag to the curb? Can Nori be trained to chunk that bag using the sensor packages - Lidar, etc to estimate range and the knowledge of the bag weight so that the optimum launch point is achievable using all those 7+1 degrees of freedom? That would be cool.

Can Nori be trained to say "This is your last one!" when fetching your Heineken? Can Nori compress to the point where it can pick up that drink tray beside that coffee table? I understand it can extend to 55" but can the grabbers reach and manipulate items on the floor, like petting a cat for instance?

In the demo videos of Nori pouring a liquid from an oat fluid container how many times did you need to adjust the level of oat fluid before Nori could execute that pour without spilling? Is Nori able to gauge the level of fluid in a container or the level of dry substance like that granola so that Nori understands the tip and spill points so that messes are minimized?

In the Teach Nori... section I can Nori being useful for managing 3D printers since the fumes in there have to be apparent and if Nori can be taught to manage printer operations that is one environmental exposure that I won't have. I worry in the clip with the paintbrush that Nori can't stir rapidly enough to clean anything from a brush. I noticed during the clip of Nori opening the 3D printer door th...

The site is broken can’t see much, but it would be great to know or show list of the components or BOM, as from experience, some companies cheap on certain sensors/motors etc.

I'm confused, is this a research device, hobby/tinker item, or consumer product?

I don't mean to be negative, but making a sustainable business on a hardware race to the bottom sounds challenging.

I'd be curious to to know from others here, what in-home tasks do people think they'll really be able to automate with any of the various bimanual or humanoid robot platforms that seem to be going around?

Humanoid on wheels might be the winning form factor.

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.

> Our open SDK includes teleoperation and demonstration tools

How is teleoperation anything more than a glorified puppet master pulling the strings or a radio controlled toy car?

Can you speak to the tradeoffs of high ratio servers vs QDD motors? Is it power or long term life?

I can see me doing so many things with this.

Can it climb stairs on a multi level house?

A robot can reasonably cost $5000, don't break your back or your health.

Why would you price your hardware so low? if this is as good as a customer would imagine, pricing should start around $20,000.

Poor buddy looks like he's got the shakes

Its main responsibility will be to stand in my front window and make the JO sign when my neighbor drives by

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.

    servo_positions = initialize()

    movement = linearmove(servo_positions, destination, 3 * second) ... // libraries will have very expressive stuff to define movements subject to limits like above
    // note: obviously this may fail, in that case DO NOT initialize movement to (0,0,0 in 0s), but instead cut the loop

    start_time = get_current_time()
    while get_current_time() - start_time < movement.end_time():
      if movement = null:
        continue // DO NOT TOUCH THE MOTORS IN CASE OF INVALID MOVEMENT OBJECT. You'll hurt yourself. Note: yourself. More so than you'll hurt the robot.
      epsilon = 0.01 // technically 0.001 should be possible. But get a decently fast controller first. Also SPI by default can't do 1khz if you're using the obvious wiring
      time = get_current_time()
      for motor in motors:
        want_pos = movement.get_motor_position_at(time - epsilon)
        send_to_servo(servo_for[motor], want_pos)
        servo_positions[servo_for[motor]] = want_pos // this becomes a problem if you're lifting stuff that's too heavy but this is a hacker news comment. Better servos will save you here

RPi5 seems a bit underpowered for all the sensors & control and offloading to a computer through wifi seems problematic. A Jetson Orin Nano would seem to be a better choice for a robot. Unfortunately Nvidia recently substantially increased the prices for their jetson devkits due to component shortages (DRAM,storage, etc), and what was $250 now costs $499 :(