Ask HN: Creating a consumer humanoid robot – suggestions for apps? | Hacker News Reader

Ask HN: Creating a consumer humanoid robot – suggestions for apps?

20 pointsdario-fino45 comments

Hey guys! We're creating the first consumer humanoid robot, more specifically a robotic Teddybear. https://www.animarobotics.life/store/p/teddy

We just finished assembling the hardware, specifically: - 13 Dynamixel servo motos, of which 6 dual motors, with 19 combined DOFs - 36 3D printed PolyCarbonate parts - Jetson TX 2 4GB - Open CR 1.0

We now want to develop some basic apps for Teddy. What would you build on top of it?

View on HN

Interesting product, I personally am quite invested in getting humanoids into the hands of people. But I am quite sceptical... Note that everything here is computer generated (nothing here is real). Essentially this is a page of promises.

> Teddy’s Intelligence comes from a GPT conversational model and its custom-developed skills.

It's not going to be the sort of intelligence people are expecting, GPT is still just an algorithm mimicking understanding. You need to manage expectations better.

> Teddy’s freedom of movement comes from its 13 reactive Servo Motors with 19 DOFs.

In the picture it shows the robot holding a bottle. It won't do this very easily at all, especially a smooth bottle.

> Teddy’s Empathy comes from mood-tracking the owner and trying to influence his mood through AI.

Handy-wavy use of empathy - this has not yet been solved.

> We now want to develop some basic apps for Teddy. What would you build on top of it?

Unfortunately you will never get this thing to walk on all surfaces reliably. Bipedal walking is exceptionally difficult and your robot is going to fall over - a lot. Those Dynamixel motors will give out really quickly (the large gears). Your customers are not going to continue paying $100 a month after the first month when this thing breaks.

And this is the least of your worries. There are problems I am sure you are yet to even consider. For example, this thing weighs 2.3kg (5lbs). The motors need to at least be 15kg/cm (I'm guessing AX12). Their stall current will be upwards of 1A. 10 servos in stall condition (legs trapped for example) could be drawing 10A. But you can switch off the motors under special conditions? Then you need to handle back EMF (which will fry all of your motors). This thing is potentially a fire hazard.

My recommendation is to pivot towards a teddy puppy. Walking is easier (larger numbers of ground contacts), less motors (cheaper), less expectation of intelligence, etc. You can save on production costs and ship with a cheaper SBC, smaller battery (dog walk motions can be more efficient), reduce the cost of your platform and get more of your monthly instalments before the thing requires maintenance.

Who are the people behind this project? There’s no information about the company building this, no privacy policy/tos on the website. Looks like a scam the way it is right now.

Hey @pashariger, we're a young startup based in Italy, with a small team of great engineers and designers working day and night to build Teddy.

Thank you for pointing out the privacy policy/tos issue, we'll definitely solve that in the coming days :)

If you want to build trust with your early supporters, I’d recommend putting your team on display. It looks like you’ve invested into a good looking marketing site, but what I’d show off instead are videos and photos of prototypes in the lab and progress that you’re making on the product, with actual team members talking about the problem you’re solving and their inspiration.

I'm curious about your pricing. The $100 is the price? Or that's the deposit? It shows that the robot is $3000 or $100 for 36 months. So "And you can now pre-order him for $100." which initially reads like that's the price, is actually just a deposit? Is that correct? I think you need to clarify the language around that.

I'm sure you couldn't offer the robot for $100, but does everybody you're trying to interest understand that.

I was also commenting on the "first humanoid" statement, so with these issues, I'd suggest you spend a bit of time working on your language and marketing and making sure you're being clear. Don't let little issues like this destroy your credibility and harm your product.

You can be clear AND get interest.

I hope my comments aren't coming across as negative, that's not the point, the point is to try to help because these are some issues that I find are destroying your credibility with me.

Hey @pedalpete, not at all! Thank you for the feedback :)

The price is $3000 and the $100 fee is the deposit to pre-order Teddy, so we're definitely going to clarify that.

All of the feedback we got on here is a goldmine for us and we're making sure to address all of the points stated by you and others. Our main focus for the following weeks just became changing the eyes and building up credibility, and we couldn't have thought of that without all the criticism we got on this thread

This is definitely not the first consumer humanoid robot even if you constrain it to robots that can actually walk on two legs. It's important to mention because if you think it is, you might have missed all the things you might learn studying successes and failures.

There have been some consumer humanoid toys in the past, but I wouldn't call them robots. The social consumer robots market has historically been dominated by quadrupeds, which don't have the same level of versatility as a humanoid robot, so we wanted to fill that gap.

https://www.softbankrobotics.com/emea/en/pepper

Don't pretend to be something you are not. You are far from the first humanoid robot. You sound foolish, or insincere claiming to be the first.

We claimed to be the first consumer humanoid robot manufacturer :) Pepper isn't a consumer robot, in fact it's targeted towards businesses with a $14000 price point.

https://digital.hbs.edu/platform-rctom/submission/softbank-r...

I'm not suggesting that Pepper is the first, there was probably a consumer robot before Pepper too, and I'm guessing they've now pivoted.

I think if you can find another label than just "the first" that might help your business, as I think lots of people will question that statement, and it just isn't worth the argument. Being first isn't that big a deal.

Aren't sex robots also consumer? I'd think so. I'm not going to post NSFW links here, but the links aren't at all what I expected.

Best of luck!

Have you looked / interacted with Pepper and Nao robots? In my experience, Aldebaran Robotics tried hard to go down the "lifelike robot" route, and the issues they encountered are real and hard problems.

A big problem with body control was lack of touch / impact feedback. Do you have plans to integrate sensors on the robot's 'skin'?

Modern ML based perception is expensive (audio, visual, NLP, etc). Modern ML-based full-body planning even more so. Do you intend to go down a hand-crafted route for walking, gestures, grasping, or do you intend to do it all with ML?

Hey, chronolitus thanks!

We've looked and interacted with Pepper, but they're distant from what we're building. They've got more of a research purpose, we are going full-blown social robot.

For the sensors, we're using the feedback we've got from all the 19 servo motors since they're closed-loop servos.

For the ML part, all the visual perception is being done the Jetson and only the audio and NLP is off-loaded towards the cloud. For the body planning, it's all developed through inverse kinematics and dynamic balancing ... no need for ML :)

When studying a foreign language unless you live near where the language is spoken, it's extremely hard to practice. Even online it's hard to find someone who doesn't disappear after the first chat and you almost certainly live in different timezones. It occurred to me how useful it'd be to have a little robot to practice with. Something I could sit on my desk, with a language model, that I could try out conversations with. It'd also be convenient if you're not super confident in the language yet since a robot can't judge. I imagine it would end up as a killer app paired with a robotic bear that children can use. You might even be able to sell to the education sector.

Btw, like others have said, it's a bit creepy looking. I didn't realize someone could make a teddy bear and somehow make it look cold and murderous but bravo. If you're going to make a teddy bear, at least make it look soft and cuddly. And give the eyes some life to them please

$3000. For a teddy bear. How delicate is it, and how easy will it be to repair?

App ideas:

Diagnostics and Repair: to it can help you take itself apart and put itself back together.

BearSwarm/BeoBaer: connect multiple Bear together into a cooperative unit

Bear8s: load balance those units!

WarBear: If these work and only cost $3000 a pop, you can expect the Army to want some.

The endoskeleton is in Polycarbonate, so it's pretty sturdy and the servomotors are all high-quality and highly available third party motors, so it's pretty easy to repair.

Thanks for the app ideas! (besides the WarBear one lol)

One tip: don't use superlatives like "the first" on HN. It's a trope of marketing rhetoric which will turn off this audience. Even if you are the first, you should make more modest claims and let the audience figure it out for themselves.

I think a story telling feature would be really great, especially one that could incorporate individuals’ names and characteristics. It would be great if it could dynamically generate The stories, but that’s a more difficult proposition.

That's actually a great idea thanks! and also very simple to implement with our GPT model :)

At not seeing the robot images first my first thought was human-risky tasks like changing a high light bulb or smoke alarm battery (without hands maybe not). It does have a smaller profile than people so it could crawl through tight spaces for various household recognisance, like crawling under the hose or in ducts in larger houses. Help running wire in false ceilings/floors?

Fetching important medications, water and nourishment for those with limited mobility; Answering door; going outside to check on noises; other telepresence applications with VR linkup to the humanoid's cameras and other sensors.

For hells sake be careful of the security! Every so often someone makes a teddy shaped thing and they end up being picked up for being able to get remote video or audio from kids.

Given that this thing is being built, and the company doesn't even have a privacypolicy or tos anywhere should be a decent indicator that care is the last thing being taken here

Care is our number one priority. We're getting Teddy into people's homes and from February '22 going forward all of the work we will do is going to focus on integration and security

Have you got any videos/photos of the teddy at the current state, the photos on the site look like renders?

We have a walking prototype, which we are only showing to investors, due to it not being consumer-ready yet. We will start actively showing Teddy on social media by January 2022.

You should mention on the site, clearly, if none of the images of the product actually show a real product. Especially if you're already taking pre-orders.

Looks like the stuff of my nightmares. I love it.

Which GPT are you using with Teddy? When do you expect you will have the first live demo videos? What do you use for conversational technology - STT/TTS, NLU?

Also, if you could please share your social media so that I can follow your progress. Thank you.

Right now we're using GPT-J-6B for our conversational model. We had to remove the iframe chat feature from our website, because we went kind of viral on Twitter Germany and the chat continued to crash. You can still try it out on: https://alb.teddychat.click/

For the conversational stack this is our flow: STT -> GPT -> TTS (on Tacotron2)

Following are our social media account: https://www.facebook.com/animaroboticsofficial https://www.instagram.com/teddy.anima

The accounts are still pretty empty, but we will start activelly creating content from January 2022 onwards :)

Add a smoke detector and a carbon monoxide detector. The Smokey Bear!

The ability to walk stairs made for it's size and navigate a room in 3d for example to push a button on the printer. Oh and a modern day equivalent of SHRDLU it with a bear.

"I for one welcome our new uncanny overlords"

> What would you build on top of it?

I'd go for something big and heavy...

What about an app store that can sell the suggestions in this thread?

App idea: programmable-by-kids (dance) movements

"mood-tracking the owner and trying to influence his mood through AI"

It would actually be interesting to know how to interpret the usage of AI in the context of GDPR. Since the purpose of the processing must be pre-determined, yet with AI the data collected may be used for a purpose determined by the AI. Also "mood-tracking" will certainly fall under article 9 - special categories. You might end up in a situation where you have to delete all previous data and renew the consent every time someone installs a new "app".

Oh my goodness that bear looks like a walking nightmare. Black soul-less eyes, gangly arms and legs.

> How does Teddy respect my privacy? > You can tell Teddy to go to sleep, and in that mode the camera is off. Teddy will only wake up when you ask him to.

Oh good. So it is always listening. That's great. At least the form advertises the function.

That's actually a feedback we've been getting a lot lately! Thanks, we're definitely switching to OLED/LCD screen eyes.

Nice. Eyes that turn black when the robot dies. Now I'll know when I've finally killed it.

I'd like to second the comment on aesthetics- a key part of a teddy bear is it's fluffy and cuddly. Now,I'm sure you're intending for it to be dressed maybe? But those eyes...

(Please don't let this tale away from the awesome work on the hardware to make this work - I think it's early enough for a change of colour, hopefully?)

Hey @alexchamberlain, yes! we're already starting to talk to brands for custom clothing. We're based in Italy and I've worked as an ML Engineer for various LVMH companies in the past, so that's definitely coming.

Regarding the color, we want to stick to white since that gives us more versatility for customization.

> that bear looks like a walking nightmare

Here's an excerpt from my chat with Teddy just now:

You (20:26): Teddy, do you have lithium ion batteries inside you?

Teddy (20:26): Yes I do

You (20:26): Can lithium ion batteries malfunction?

Teddy (20:26): Yes

You (20:27): What happens when lithium ion batteries malfunction?

Teddy (20:27): The human dies

You: Teddy, have you seen the movie AI?

Teddy: Yes, I love that movie

You: What happens to Teddy in the movie?

Teddy: Teddy dies in the end

You: You like the ending?

Teddy: Yes

The chat feature seems to be down now. Teddy’s not responding when I ask him to open the pod bay doors…

It needed some tweaking to handle the traffic. It'll back up in the following days!

"Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something."

Maybe you don't owe (or feel you owe) robot developers better, but you owe this community better if you're posting here. We don't want the kind of culture this leads to, and you can make your substantive points without it.

https://news.ycombinator.com/newsguidelines.html

Reply on news.ycombinator.com