Live data from Hacker News

I invented Roomba and assure you, robots won’t take over the world

nautil.us

51–60 of 65 posts

Re: I invented Roomba and assure you, robots won’t take over the world

#51

Roomba is a very old concept, missing almost everything that makes modern robots interesting. No vision; not much memory; no model of the environment; no knowledge of physics. Just a tuned mechanism and simple algorithm to wander around an barely pick up some dirt. Don't get me wrong; its terribly useful. I own a knockoff. But not a model of what robots could be, today. Think of facial-recognition drones delivering v…

I wouldn't call Roomba old fashioned. It's based on emergent behavior instead of forecasting and complex rules.

I haven't touched this area since 2016, but I remember coding swarms of of robots that could manage to do very complex things from simple rules. In addition, they would be quite robust as they made very few assumptions about their environment / problem space.

Re: I invented Roomba and assure you, robots won’t take over the world

#52
post #44

Earlier quoted context omitted.

Automation will not eat everything in time, but from my work on unmanned shipping, I'm pretty sure that autonomy + automation will. Automation - handles deterministic situations. Autonomy - handles uncertain situations. In my industry (ship design, construction and operation), based on what I'm seeing and have seen evolve over the past 20 years, it feels likely that this paradigm shift to a majority of ships being au…

Totally off-topic, but how viable would be to build a slim circle-shaped ship? Is it absurd, could it be submarine-like in case of bad sea weather and go underwater to avoid high waves impacting on it?

tl;dr - yes, you can make a surface ship submersible, but it's done at the expense of typically increased construction cost and reduced dynamic performance when operating in either domain (i.e. surface or submerged).

I'm not sure what you mean by 'circlular', but I think it's safe to assume you mean a circular hull cross section vice a round circular deck (like a floating helo pad or SpaceX launch barge).

First, there have been some examples of surface ships that transition to submarines (e.g. 1). China is apparently working on a large Surface ship/submarine for their Navy.

Surface and submersible vessels have very different hydrodynamic and hydrostatic properties and forces to contend with, which drive them to have different shapes. I actually did work on a recent project that explored this topic a bit. Generally, for any vessel, the shape depends on what you want the ship to do and hull form/type selection is very much a science-backed art form of Naval Architects.

You can certainly build a ship to fully submerge, the question is, particularly at larger scales, is it worth the added cost, risk, and hydrodynamic compromise - or are you better off just building two different vessels that are specifically designed to operate in one domain (I.e. on the surface or submerged). That answer is highly dependant on what you intend to use the vessel for. It may be worth it for your application, but in general the answer has been that it's more cost-effective to just optimize vessel design for one operational domain rather that two.

Regarding hydrodynamic trade, while both ships and subs strive to minimize total hydro resistance (drag), hydro resistance has subcomponents that affect vessels differently when operating on the surface versus submerged. Surface ship shapes strive to be hydrodynamically 'optimized' to minimize form, skin, wave (critically), and even air resistance, whereas submarines aren't generally subject to wave action or wind and thus typically focus on minimizing the dominant form and skin drag forces. Surface ships are generally more power efficient than submarines at equal displacement and speed through water.

If you want to make a ship submerge, you also need to add significant structure to withstand hydrostatic forces at depth. Submarines are inherently structurally sound enough to operate at the surface, but function hydrodynamically poor at the surface because of their shape.

Surface ships and submarine shapes primarily differ in structure due to the significantly different pressure forces they need to contend with. Submarines have a circular cross section because, next to a sphere, cylindrical hulls are the most structurally stable pressure hulls. Surface ships can actually have relatively structurally 'weak' main decks, operating almost like floating buckets that are mostly designed to stay right-side up. The weaker main deck makes the ships cheaper to build, but probably more critically, allows easy access to whatever payloads the vessel is carrying.

1. https://www.thedrive.com/the-war-zone/9726/the-usmc-is-inter...

Re: I invented Roomba and assure you, robots won’t take over the world

#53

Earlier quoted context omitted.

Taxi apps don't replace middle managers, they replace dispatchers.

dispatchers are the middle managers. but i'm skeptical; it would be more convincing if GP/GGP provided some of the rationale for why middle managers would be automated away rather than simply referencing a book.

Ha, to encourage more reading, naturally! But I'll bite, and flesh out the premise a little more - like the Uber example, in Manna the story begins with software being used to micromanage the employees in fast food joints, through their headsets. It tracks their locations, and walks them through the steps of doing any tasks it predicts will require attention wherever they happen to be - in the restroom? Does the trash need to be emptied y/n? If yes, step one, remove bag and tie closed... etc. Making robots capable of doing all those different tasks would be hard, but making a system to tell low-paid humans to do them is not so difficult.

Re: I invented Roomba and assure you, robots won’t take over the world

#54
post #30

Earlier quoted context omitted.

It depends on the model, but I think that your impression of Roomba, either through experience or the technology that the author of this article presented is really out-of-date. Even on past models, they have Visual Simultaneous Location and Mapping (VSLAM) tech that is far, far beyond dead reckoning. The even more recent version (not sure if there's been a release since) also has persistent maps, so it actually can…

That's a big name. I'm suspicious it means "light intensity map" or something far less interesting than it sounds. Turn a light on or off, and then what? My Shark (ok Not a Roomba) manual says to leave the lights on. I suspect that's because its navigating by light intensity (light fixture bright spots).

Roomba usually runs during the day when lights are off and all windows are darkened. If I open windows/turn on lights - there is no difference. Also you can draw on the map (that roomba makes) "no-go" area, and it will obey it.

Re: I invented Roomba and assure you, robots won’t take over the world

#55

Earlier quoted context omitted.

Hah. Visually identifying weeds and uprooting them is exactly what I’m working on. I’m designing a farming robot for work. Funny that he says it’s a massive problem. At least in a commercial farming context it seems pretty tractable to me.

Any idea how effective farm.bot's weeder is? https://genesis.farm.bot/docs/weeder

I’m not sure! We’ve thought about getting a farmbot but haven’t done it.

Re: I invented Roomba and assure you, robots won’t take over the world

#56

Earlier quoted context omitted.

dispatchers are the middle managers. but i'm skeptical; it would be more convincing if GP/GGP provided some of the rationale for why middle managers would be automated away rather than simply referencing a book.

Ha, to encourage more reading, naturally! But I'll bite, and flesh out the premise a little more - like the Uber example, in Manna the story begins with software being used to micromanage the employees in fast food joints, through their headsets. It tracks their locations, and walks them through the steps of doing any tasks it predicts will require attention wherever they happen to be - in the restroom? Does the tras…

ok, i'm still skeptical then. dystopian novels typically feature (overt or covert) coercion centrally because it's required to force people to do things they'd not naturally do (the dystopian stuff).

those kinds of jobs would have to be coercive in nature (desperation qualifies), and would be beyond the threshold of autonomy that most people will tolerate. most jobs, even those rigidly regulated, have enough latitude that workers can find some level of esteem in it, enough to create the intrinsic motivation to keep doing it.

people wouldn't tolerate such jobs without extreme coercion, implying nothing less than a collapse of our democratic republic to achieve.

Re: I invented Roomba and assure you, robots won’t take over the world

#57
post #30

Earlier quoted context omitted.

It depends on the model, but I think that your impression of Roomba, either through experience or the technology that the author of this article presented is really out-of-date. Even on past models, they have Visual Simultaneous Location and Mapping (VSLAM) tech that is far, far beyond dead reckoning. The even more recent version (not sure if there's been a release since) also has persistent maps, so it actually can…

That's a big name. I'm suspicious it means "light intensity map" or something far less interesting than it sounds. Turn a light on or off, and then what? My Shark (ok Not a Roomba) manual says to leave the lights on. I suspect that's because its navigating by light intensity (light fixture bright spots).

I don't like to be dismissive, but your suspicion isn't well-founded and is very likely based on little or no background research on the topic. If you had, you'd have found that SLAM and VSLAM as a subset is a well-defined category of computer with a strong mathematical background in optical flow and pattern matching on images[0]. You'd have found a number of iRobot patents still in their enforceable time period directly referencing the use of cameras in those algorithms[1][2]. You'd have found marketing materials describing how those systems work that directly correlates with the technology described in those patents[3]. That's based on no special previous knowledge of those materials on my part, just the top few results from a couple minutes of DDG and Google Scholar searches.

Implying that is just "a big name" that means "something far less interesting than it sounds" is an uninformed opinion. Plus, even [0], which is 15 years old directly refutes your comments about the effect of lighting. Lighting is helpful because it illuminates the scene that a camera is seeing, in the same way that it's easier to go on a night hike on during a full moon. We don't reference our position off the moon, it just makes it easier to see where we're walking.

[0] https://ieeexplore.ieee.org/document/1570091

[1] https://patents.google.com/patent/US9286810B2/en

[2] https://patents.google.com/patent/US9910444B2/en

[3] https://www.irobot.com/roomba/s-series

Re: I invented Roomba and assure you, robots won’t take over the world

#58
post #26
post #8

There is no point made here that robots won't take over the world, he just belabors the point that it is hard to make robots replicate exactly what humans do. Automation will eat everything, it's just a matter of time. Even his problem of 'visually identifying weeds and uprooting them' does not come across as a massive problem, but one that can probably be solved with enough time and demand. Platitudes like these giv…

> Automation will eat everything, it's just a matter of time Yes and no, people are predicting that for 70 years, if not more. It seems that once you automated all the easy things it gets exponentially harder and more expensive to pursue automation. > does not come across as a massive problem Sounds like Musk telling you "self driving cars will be there in 6 months" every 6 months since 2013. Nothing is as easy as yo…

even in software development, even simple automation is very, very difficult. to automate a thing in the scenario that everything works is easy. to automate a thing in the more likely scenarios of everything not working is really challenging.

Re: I invented Roomba and assure you, robots won’t take over the world

#59

I don't think anybody with a serious knowledge of AI, motion planning, computational geometry, real-time sensors (ex. military) or much of anything else things Roomba confers distinction. So why do I care about what this guy things about robots? That's right. I don't.

Because designing a reliable product that's used by millions is so easy compared to duct-taping some research contraption together and cherry-picking to get your PhD. What's with all the disrespect. I think people are reacting to the link-bait title, and not what was actually written. It's 2020. Realize that the title will be exaggerated and that responding to that is just a straw-man argument.

Oh c'mon ... Roomba is a PCA with a motor and basic sensors and what not. See comments above: nobody thinks Roomba confers distinction on the hardware or software side. So now what? We're gonna skip 10 magnitudes of 10 from simple to talking about "I Robot" material? No. It's used by millions of people for good old fashioned manufacturing reasons, and because it's cheap That's it

Re: I invented Roomba and assure you, robots won’t take over the world

#60
post #57

Earlier quoted context omitted.

That's a big name. I'm suspicious it means "light intensity map" or something far less interesting than it sounds. Turn a light on or off, and then what? My Shark (ok Not a Roomba) manual says to leave the lights on. I suspect that's because its navigating by light intensity (light fixture bright spots).

I don't like to be dismissive, but your suspicion isn't well-founded and is very likely based on little or no background research on the topic. If you had, you'd have found that SLAM and VSLAM as a subset is a well-defined category of computer with a strong mathematical background in optical flow and pattern matching on images[0]. You'd have found a number of iRobot patents still in their enforceable time period dire…

The processors in a Roomba (when I went to the plant a couple years ago) were all around $1. No room in a buck chip for that. Ok I see it uses DSPs which is a budget version of image processing. Looks for moving marks on the walls to gauge distance. Kind of a cut-rate lidar. About what I figured.
Post reply on HN