Ask HN: What would it take to make a robot vacuum in the USA?

13 points by pc2g4d 1 year ago | 18 comments

There is no robot vacuum manufactured in the United States as far as I can tell.

It seems crazy to me---these devices map out the interior of people's homes and businesses, often with cameras running and a wifi connection to the Internet. Talk about an espionage giveaway... all to a certain authoritarian country in Asia which manufactures 95% of the devices.

I used to be a big free trade advocate but have come to feel that free trade is something of a gift, which should be given mainly to countries whose vision of the world is compatible with ours. It also should not be allowed to undermine security-critical technologies and capabilities, such as electronics and microprocessor manufacture, etc.

In spite of recent tariffs and moves to restrict exports of sensitive technologies, the tech ecosystem in the US seems to be in a seriously impaired position.

If one were to push back against this and try to launch a robot vacuum product assembled in the USA of US-made components... what would it take? What are the economic impediments to doing this in a profitable fashion? Is there an economic policy that would make this feasible?

It seems there are a number of key components: * Microcontroller * Sensors - laser or lidar to map environment - something to determine floor type / depth * Bluetooth antenna to interface with a mobile app * Vacuum * Dust compartment * Dock to autoclean * Embedded software * Mobile app

Such a device would be an ideal candidate for training using reinforcement learning. (Has this been tried? Or is "robot vacuum" already a solved problem?)

My training is in software and machine learning. Yet it is in on the material side of things where the U.S. lags the most. There are so few physical things of relevance to modern life that we are making here these days, and even fewer at a competitive price point.

Would love to hear others' thoughts. Thanks

EDIT: A relevant article from 2011: https://www.forbes.com/sites/stevedenning/2011/08/17/why-amazon-cant-make-a-kindle-in-the-usa/

talldayo 1 year ago |

There are local-only robot vacuum options: https://valetudo.cloud/

But you shouldn't ask yourself why it isn't possible. The question is why it's not successful, or why there's no demand. America is satisfied by foreign alternatives. We don't want secure devices, we want cheap and simple devices that come in pretty boxes with the labels we like. Roomba et. al wins because it caters to the lowest-common-denominator, the basest American that most shoppers embody.

And America, collectively, cannot care. You won't see a law demanding software transparency or firmware alternatives for domestic robots. We won't assemble them locally because our labor is too expensive, we won't manufacture the chips domestically because it's cheaper to import them. You won't see people owning USA-made Roombas for the same reason you don't meet people using a Purism phone instead of a Samsung/Apple handset; the market has spoken, and they are absolutely apathetic to anything but price.

pc2g4d 1 year ago | |

It appears that Valetudo is an alternative firmware for existing robot vacuums. The OpenWRT of robot vacuums if you will.

I don't think it's accurate to say that it is simply us wanting cheaper things. There has been a deliberate trade and even monetary policy that has led to this outcome. It is a matter of policy that the expertise to manufacture such things no longer exists in the United States.

For example, the WTO trade rules the U.S. engineered and China's entry into the WTO facilitated the outsourcing by which managers of U.S. companies sold core industrial competencies out.

Another policy has been the maintenance of a strong dollar, bolstered by the petro dollar system. This makes U.S. exports on the whole uncompetitive.

The abandonment of vocational training in favor of a college-for-all approach has also undercut the skillbase that would be necessary to maintain a domestic electronics and devices industry.

In essence, you could see it as an abandonment of the working class in favor of college graduates, whose stock portfolios benefitted from the cheaper costs accessed in the Chinese labor market.

The America you are familiar with perhaps doesn't care. But many Americans understand that the trade and monetary policies have undercut their own earning potential, and they buy those cheaper Chinese goods with great resentment, simply because _there are no longer alternatives_ in a huge number of categories.

Eventually the dollar will lose its "reserve currency" status; eventually enough of the rest of the world will be developed enough, or trade relationships fraught enough, or free trade discredited enough, that the pendulum will swing the other direction. I suspect the swing has already begun. Part of my motivation for writing the original post---to test the waters.

talldayo 1 year ago | | |

> The abandonment of vocational training in favor of a college-for-all approach has also undercut the skillbase that would be necessary to maintain a domestic electronics and devices industry.

This is just wrong; it's literally the opposite problem. The United States has been raising kids on 1960s curriculum for half a century, and now the only workers we can produce as a nation are the ones being trained and paid cheaper in China. College-for-all is the only policy that works; the United State's economy, not monetary policy, has ensured that our labor will never be competitive unless we tax imports. And if you tax imports, you might as well be admitting that the US can't compete anyways.

> But many Americans understand that the trade and monetary policies have undercut their own earning potential, and they buy those cheaper Chinese goods with great resentment, simply because _there are no longer alternatives_ in a huge number of categories.

Lmao. Yeah, with "great resentment" the average American buys their iPhone. There's just no alternative!

No, the average American (and most of the "good old boy" conservatives you're describing) care about nothing other than price and marketing. They'll piss and moan about oursourcing while buying a Vietnamese slim-jim in line to pay for Saudi-enriched fuel at the gas station. They'll shake their fist and claim it's the government's fault, but they know full-well a lassier-faire economy would only exacerbate the problems of foreign influence. China and Russia do not somehow become less equipped to manipulate our economy with less regulation. That's a fantasy.

You know what? Go build your robot vacuum. I'm actually really moved by your plight, and nothing would make me happier than watching you retread this perennial problem. Just don't come crying to me when you spend more than the MSRP of a Vision Pro paying for 4 locally made brushless encoders. We tried warning you, you insisted otherwise.

bediger4000 1 year ago |

"Robot vacuum" is not a solved problem. I've got a Roomba, and it mostly does the vacuuming, but sometimes gets "stuck" in the middle of the room, or maybe against a single wall. My Roomba's nemesis is chair legs, sometimes it gets under a 4-legged chair, but somehow can't find it's way back out. Comical, really. I enjoy my Roomba as something of a pet, but I have come to low expectations about vacuuming completeness.

pc2g4d 1 year ago | |

If one could, say, solve this problem, which is apparently quite difficult, then it could be the basis of a premium product?

GianFabien 1 year ago | |

Same software as self-driving cars?

iamflimflam1 1 year ago |

Software guy looks at hardware and thinks “how hard can this be”

pc2g4d 1 year ago | |

No... I ask this question in all humility... but also frustration---my ignorance of hardware is part of my country's ignorance of hardware, because we don't make things here anymore, broadly speaking.

iamflimflam1 1 year ago | | |

One way to think about manufacturing something is to see where the line is between assembly and manufacturing.

With electronics you might get your PCBs manufactured and assembled by someone else (it’s a specialised job that needs expensive equipment).

But you can do the design locally with skilled engineers.

With enclosures and plastic parts you need injection moulding setups - again a specialised thing to do. So you might also choose to outsource that.

But you have local people who do the CAD and mould design.

What you’ll end up doing is the assembly side of things - so someone else will be doing the actual hard manufacturing of parts and you’ll be plugging things together and packaging things up.

As soon as you realise that then it’s quickly obvious that should be done where all the component parts are being manufactured.

So, you end up just doing the R&D part.

But eventually you realise that your manufacturing partners understand and know a lot more than your local people.

So, you end up outsourcing everything to China…

BobbyTables2 1 year ago |

Getting rid of carpet would also avoid a lot of the need for a vacuum in the first place. A simpler dust mop would then suffice.

hnthrowaway0328 1 year ago |

Getting America back to manufacturing is an expensive and long term job. And it probably requires some bloodshed. No one os going to do that.

The next best thing is to source from countries that is not China or whoever you don't like. There are still cheaper, maybe not so cheap options in other countries. SE Asia and Eastern Europe come to mind.

b20000 1 year ago |

if this was entirely manufactured in the US the retail price would be $10,000

pc2g4d 1 year ago | |

U.S. manufactured microcontrollers can be had for < $100 for instance. With the parts, the assembly per unit must take on the order of minutes. How do you get to $10,000?

talldayo 1 year ago | | |

Chinese manufactured microcontrollers can be had for <$0.10, bought in bulk.

I used to work on an FIRST robotics team in high school; every year we had to fundraise ~$20,000 for the component cost (not labor for software or hardware) of our USA-made robot chassis. Our robots were bigger, but that really only amounted to a larger aluminum frame. And just so you know, our assembly period took around 6-8 weeks (and we didn't have to miniaturize anything).

If you insisted upon making everything here in America, I legitimately would not be surprised if you ended up with a multi-thousand dollar component bill for each model sold. The reason why China beats us is because they can mass-produce everything needed for the supply chain of these machines, and we cannot.

b20000 1 year ago | | |

distribution of consumer electronics can mark up the price with 30%-50% and then the retailer adds another 30%-50%, and then the cost of transporation needs to be paid, there is the cost of storage, sales tax, etc etc

so a $1,000 MSRP consumer electronics item might need to be manufactured for a cost of $100-$200.

i think that if one wants to make products in the US, the cost of distribution and retail and transportation needs to be addressed, and it seems nobody wants to touch that hot potato

theGeatZhopa 1 year ago |

It's one thing to have the parts designed and moulded, the electronic PCB boards to be made, the software to be written, assembled and packaged. The other thing is storage, warranty, distribution. But also selling and advertising should be kept in mind.

Up to here, it could be done as a one man job. If you're a multitasked and multitalented in using all the software for design and production. Typically it's not a one man show. So how much people are needed to be paid in such a local company?

But then, it goes to the regulatories. FCC approval. Some states special regulatories. Want to sell it world wide? .. Europe? .. EC sign, even more regulatory stuff like upfront waste-fee, etc.

What about the licenses and patents? Even the app communication with some devices can be problematic in terms of patents..

So, parts design, parts production, parts assembly, packaging making and packaging of the product itself, is not that big problem and can be done locally or outsourced it locally.

Also the whole regulatory stuff and danger analysis and what ever is needed before you actually can sell to customers, can also be done or outsourced locally.

... Just some money is needed for it. Your friendly local outsourcers also want to earn some money for a living.

So you end up with big development costs per device. No one will buy it for that costs.

That's why you typically search for partners, who have some experience in rapid development and prototyping. When you've found them, they also experienced in final production and packaging. And because it's their job, they often also know about the regulatory laws and/or are experienced here and use f.e. less plasticizers in the rubber..

Most of the times you'll find such partners in eastern/asian world. Because the wages there are lower, so a company can exist even if it doesn't make own products.

It's not the right question you ask. The right question would be "how expensive will it be to develop and produce locally". Because it's all about the money.

(I saw in a comment of you further down "us manufactured microcontrollers < $100"... That's exactly the wrong approach :)

Compare it with the costs somewhere else. The aim is to preinvest as less money as possible and to earn as much as possible per product. Don't forget all the other costs that will be pile up and must be turned over per device. So you might really end with 10k-20k per device. How much devices are planned to be sold in year in US? )