I made a big one for my garden. It is like a that scene in honey i shrunk the kids with Rick Moranis laying in that cantilevered flying cot thing but i use 2 high speed winches and some machined aluminum and the cabling sort of like an etch a sketch controls. I control it with the same usb gamepad that titanic sub imploded with. i can raise and lower myself about 4 feet to get inbetween the rows. It is pretty fun to play in too. I want to build a robot to get in it and then remotely control the whole thing. Its basically the same thing as the NFL spider cam but holds more because I'm a fatass and performs much worse. I used 12 foot 4x4s with concrete bases. Its about a 20 feet by 35 feet garden. My only real concern is the longevity of the cables, the hight im suspended at isn't really dangerous but i know i am not using the right cables for the winch. 3/16 in marine grade vinyl coated 4200 pound working load limit. my spools or whatever you want to call it is 6 inches im afraid theres too much tension on parts sometimes. I had an alternate design that used water as a ballast and you just pump or drain water to move around the grid but that was too ambitious for my lazy ass.

Hi there, I'm Nathaniel, the creator of this robot. Thanks for posting it :) If you have any questions feel free to ask.

Can it kill you? no probably not. The tension limit kicks in and lets the line out if you pull too hard.

Will it work in a room with a ceiling fan? usually its fine since you can lower the work area ceiling to exclude the fan.

Will it make my kids learn to not pick up after themselves? They never will anyways.

> Tension is all you need

Rarely does a product announcement make me laugh out loud :)

(https://en.wikipedia.org/wiki/Attention_Is_All_You_Need)

My thoughts while watching the video

Robot picks up clothes -> this is stupid

Robot picks up kids toys -> this is brilliant

In reality though, I think the cables will never work in a household situation. It's a interesting concept and I'm sure it was fun to build :) Makes me wonder what commercial application this would work for, I'm sure there's something out there that would benefit from this over waiting on humanoid tech

  Australia's $900-million-a-year greenhouse tomato industry currently relies on farmers pollinating by hand
    “So we've managed to design a robot that pollinates greenhouse tomatoes.”
    “It autonomously drives along the rows, finds flowers that are ready for pollination using artificial intelligence, and sends air pulses to vibrate the flowers in a very specific manner to imitate buzz pollination as performed by bumblebees.”

Awesome!!! I love projects like this.

Reminded me of the Hangprinter https://github.com/tobbelobb/hangprinter which has been around for around for long time... always wanted to build one.

ETA: https://hangprinter.org/

I kind of want to try something like this for garden weeding. Put a small spinning blade instead of a grabber arm on it. Computer vision can identify any weeds popping up and it'll go over and cut them before they get too big/woody for the blades. The challenge would be dodging taller plants and probably the forces that the blade would put on it.

To those asking about more demanding tasks, I totally think this thing could scale up, reminds me of Maslow CNC (and predecessors):

https://www.maslowcnc.com/

basically a cable driven 4x8 sheet router.

This really needs a way to pull the strings and bot out of the way when not in use.

Something like an extra string running along the edge of a wall that could “gather up” the diagonal strings and tuck them away on an edge or corner would make this a much easier impulse purchase.

Very clever - and I love "tension is all you need"

Can it make a bed? Actually, can it lift a comforter? Can it fold towels?

I think the form factor is actually fine but the specs probably need a boost.

I love the ingenuity, though it's all fun and games until someone or something manages to yank down on one of the cables...

Do you think it can be scaled down to pick from random pile of M2-M5 bolts and sort into Gridfinity bins?

I was actually just thinking about something like this this morning. Mainly because I was idly wondering if the kinematics would work with just 3 supporting cables. Intuitively, it seems like it should, but I haven't bothered to research it yet.

Anyone?

I love this so much.

I built a big swing arm in my dorm, so I had two 40lb+ 21" CRT dangling free under some pulleys. Keyboard mouse space below. Then myself on a hammock chair, also swinging. Just a couple pieces of 2x4 reinforced with some angle iron, some door hinges, some cables & pulleys. It was really fun.

I've dreamed of having just a good 2d robot that can, like, hold a camera and pan. I saw Stringman here a couple of weeks ago & it's been hard to not thing about it! I don't own though, so, seems unlikely I'd commit, as alas the walls are not a simple white I can easily patch & paint. But so cool!

I've been wanting to build a cable robot for Halloween which flys a ghost around the front yard with a vision system that detects people passing by and the ghost floats towards them and follows them.

Awesome! Though I'm thinking that you can do MUCH MUCH more things with 2 grippers than with one. Has all been tried in terms of weight management to instead of having a motor in the air with a gripper, putting two motors on each end corner of the ceiling that can pull a line tethered to a (in the air) pulley system that grasps? You could then have say 4 lines to pull (1 left grip, 1 right grip, 1 to adjust both gripper's distance, 1 to rotate the in the air platform).

It's very (in a good way) b movie scifi. bonus if it makes 'wacka wacka wacka wacka' sounds can see how this could be usefull for theater, advertising, dyoramas, promotional displays, where the kind of chunky movement will be a feature and not a detractor in our otherwise hyper realistic but janky digital lives

It's an interesting concept. If you scroll through the page you'll see it's actually pretty strong. The blanket pick up video should actually be the first video.

My fear is that a company that has the money to build and manufacture this will rip off the idea and use their resources to polish the concept. With a lot polish, this type of thing can be useful in certain environments.

I was working on something similar, stoked to see this type of project making it out of the labs! The inspiration for most similar systems are industrial settings with automated bridge cranes which are fantastic, almost surgical when everything is symmetrical in the active environment.

It might be more expensive, but could this be made "inside-out", as in putting the movement motors inside the arm itself, and the walls just having cable anchors, and maybe a tag that the arm can use for position tracking?

I like that this doesn’t require giving up human living space for another machine.

It hangs above the head — the space that is mostly unused. It pops in only when needed, and stows back otherwise. It’s also omni-present within the room.

This is how Skycam, the camera system for sporting events, works.[1] Four cables, four winches, coordinated.

[1] https://www.skycam.tv/

Repo link: https://github.com/nhnifong/cranebot3-firmware

Very cool, might have to try to rig something like this up in the workshop side of our basement, which is almost always a mess.

$800 is already super cheap, and the openness makes it likely you could implement even cheaper options yourself.

I didn't understand if exactly this feature is available, but I think it would be a good idea to let the strings actually come off each motor completely when yanked hard. Interesting idea :)

Super smart but it's one of those ideas that hopefully gives someone else in this space another idea that leads to something that could mass scale and I would buy. Love the ingenuity though.

They designed the cables to go down into the room, instead of from the ceiling - seems like that is a deal breaker for most environments actually, because you need the whole room to have open space just to drop it to the floor all over.

They say they use 4 actuators, but if they only used 3, this problem would be solved.

Instead of independently controlling all four lines, you can join the lines leaving each corner unit and then just move the line left or right instead of reeling two different lines in and out. You still use the corner eye loops, and you use two cables that are X+Y in length. This allows you to pick a point on the ceiling. Then a third actuator and line in the gripper itself could lower itself to the floor.

They compare it to a robotic arm, but they are a natural combination. Two arms, even, floating above it all, could scratch the dog, make breakfast, and maybe someday, fold a fitted sheet.

This is really cool!

Would more taut crossed wires and a descending line for the grabber (like the one at an arcade) solve the humans crossing wire path problem? Then it becomes separate x/y and z.

It's an interesting concept, but I think I'll hold out for the full-body humanoid version.

There is an undeniable allure to the idea of a bipedal chassis realizing I'm an inefficient drain on corporate resources, then stabbing me to death while I sleep due to a minor Terms of Service violation.

How is the wife acceptance factor of this thingy?

Very cool and I'm excited to see commercial domestic helper robots that run locally without a subscription.

The first thing that sprung to mind is using one of these to clean up my dog's poo from my garden.

You know this is a great project but seriously just tidy your room instead.

This guy does well on Instagram. Good hustle hope it works in irk reviews. Hope it survives to get to v3

Seems to work for people who basically have very empty rooms, and very orderly piles of things.

I was thinking about one of these to find the best cellphone reception in a room.

Next level is 2 of these working in sync to manipulate objects in the air.

When not in use it should just hang on the ceiling as a fancy lamp

Reminds me of the surgery robot. Talk about an unbeatable surgery conversation:

"Make another incision ... NO! QUICK! Plug the hole with the knife"

https://www.youtube.com/watch?v=A_BlNA7bBxo

Or perhaps I should say: what about safety? Not that any of the billionaire AI companies could have been created if they actually cared about (legal, mostly) safety ... so perhaps you should just go ahead and ignore it.

You should make a demo where it plays chess with an actual board.

Could it put something in a cupboard?

Looks good, but I am waiting for Droneman

CN you send me any real info on the robot you built it's hard to comment on the things I know nothing about.

it shows it picking up toys but imho this looks dangerous for kids.

could get tangled up in those wires...

My cats will love that :)

That's awesome!!

get funding for this, it's worth a shot

Crazy expensive

wait how many cameras do you need?

Come on this is not up to the standard. Consumer product must be highly polished, it can be fully unfuncyional, but it needs to look good. This one is close to my 8 years old son's tape+random lego pieces...

LOL @ tension is all you need

Looks jank as fuck. Why can they not stop the wobble?

At least it’s not a straw man! I will see myself out

I wonder if this idea can be commoditized /s