[ / / / / / / / / / / / / / ] [ dir / animu / baaa / choroy / dempart / monarchy / tingles / vg / vichan ]

/robowaifu/ - DIY Robotic Wives

advancing robotics to a point where anime cat grill meidos in tiny mini skirts are a reality


Scheduled downtime for server maintenance:
April 25 at 12:00 noon PST

March 2019 - 8chan Transparency Report
Subject *
Comment *
Verification *
Password (Randomized for file and post deletion; you may also set your own.)
* = required field[▶ Show post options & limits]
Confused? See the FAQ.
(replaces files and can be used instead)
Show oekaki applet
(replaces files and can be used instead)

Allowed file types:jpg, jpeg, gif, png, webm, mp4, pdf
Max filesize is 16 MB.
Max image dimensions are 15000 x 15000.
You may upload 5 per post.

board irc: freenode  #robowaifu

File: 04b4a71b49b85d1⋯.jpeg (228.11 KB, 1024x768, 4:3, chobits2.jpeg)


Why Robowaifu?

Modern women have failed their men and their societies, and men around the world have been looking for a solution. History shows there are cultural and political solutions to this problem, but we believe that technology is the best way forward at present – specifically the technology of robotics. We are technologists, dreamers, hobbyists and geeks looking forward to a day when any man can build the companionship he desires in his own home. Not content to wait for the future, however, we are bringing that day forward. We are creating an active hobbyist scene of builders, programmers, artists and designers, using the technology of today, not tomorrow. Join us!

>We are also on Freenode IRC at #robowaifu

>FYI your real IP is visible when using IRC. Consider using a VPN.

Note: This is generally a SFW board, given our engineering focus primarily. There is a NSFW board >>>/clang/ that would be a suitable choice for those so inclined. More bizarre NSFW is available at >>>/monster/.

What is a Robowaifu?

That's a very good question! A robowaifu is a robot that fulfills a man's natural need for companionship. To achieve this, there are many problems to be solved, and there are many approaches being explored here. Solutions range from silicone sex dolls with robotic features to VR and virtual girlfriends, and even into the realm of spirituality.

Do You Intend to Sell Robowaifus?

The objective is to design a robowaifu that any man can create in his own home. We also believe in open source as a way to protect our rights. We expect that some day companies will sell robowaifu kits. Already there is one company that sells a complete robowaifu, that company being RealDoll, the robowaifu being named Harmony.

Post too long. Click here to view the full text.

Post last edited at

File: e104560c020b84c⋯.jpg (66.21 KB, 736x552, 4:3, 4a9d3d2d53d3367b91f902a60f….jpg)

File: 3ca0172b8d592e7⋯.jpg (63.91 KB, 650x434, 325:217, holo cardboard cutout.jpg)


To keep this board just alive, I will chronicle my development of my first waifu-type bot. In the Fluffytail thread I realized I projected too far ahead with final character designs and 3D models, when I hadn't even completed a single decent robot at the thread's creation. Now that I have experience making the basic robots I feel confident enough to tackle this next challenge. Besides, before I make a catgirl I better make a mousegirl first.

If for any reason 8ch goes down, I will start a Twitter account and will follow the you-know-who's of Japanese robowaifu twitterati which have been posted before, I'm just not into a social media mood right now because of all the virtue signalling.

Why Micromouse?

It is small and mechanically simple, while at the same time it has very precise movements which give the illusion of intelligence. I want to make a tabletop pet similar to Anki Cosmo or Vector. I was originally going to try to make a walking Rachnera quadruped but it's more complex with walking robots having to factor maximum weight. With a micromouse the only constraints are length and width to be able to fit into and traverse a maze.

Minimum Viable Product Robowaifu?

This is just going to be a basic pet that runs around erratically, makes beeping noises and lights up some blushing LEDs. It's going to be pretty tall, not your typically low profile micromouse, and to make it a waifu it will merely have a cardboard cutout of an anime character propped up on top of the chassis. An infrared sensor will be placed on the head to determine whether a hand is simulating a headpat. It will have multiple modes so that it can behave like a typical micromouse as well as turn into a waifu.

Supercheap Opensource?

Unlike University / Competition-grade micromice that you see from Japan/Taiwan, this is just going to barely qualify. The parts total should cost no more than US $50, with no single module costing more than $4. Many parts are from grab bags or off-the-shelf cPost too long. Click here to view the full text.

17 posts and 23 image replies omitted. Click reply to view.


File: daf78dc29a02a36⋯.png (185.5 KB, 2928x1409, 2928:1409, TCRT5000 LM393.png)


output slammed by LM393 or LM339 comparator so the signal can be properly pulled as close as possible to Vcc or ground to give a proper digital signal and also enough to drive status LEDs.

Now I don't know whether it's my soldering, but I've tested on two different pro minis and inputs A5 and A7 seem to be susceptible to oscillation when there's comparator load on the other inputs. Though I haven't tried on a Nano yet (the whole point of the Pro Mini is that it's smaller and doesn't need bootloader). I'll also try taking out all comparators and feeding all infrareds directly as analog inputs as levels 0 through 1023. Comparators are needed to drive the status LEDs so I won't have my christmas lights, but this time I'll prioritize functionality and ease of wiring.


File: 0d2e915ff4f4a4a⋯.png (18.68 KB, 1141x834, 1141:834, opti.png)


>All that analog circuitry

Have you tried something like this yet? The sensor half of your IR LED thing there is a phototransistor, meaning that the current flow through it increases with received light intensity. Reading the voltage before the phototransistor will give you a usable distance reading. The circuit in my image should give you 5V at 10+ cm (no bounced light) and the voltage drop of the phototransistor at full intensity (~0.7V or so, you'll have to experiment with distance and reflectivity to get the minimum distance) with a more or less linear scale with distance between those two. However, your reading will likely be overwhelmed with background noise. To eliminate that, apparently it's common practice to pulse the LED, on for T ms, off for T ms. You read the analog voltage during the off phase, which is all background noise, and subtract that from the reading you take during the on phase to get the true distance. Accuracy increases with smaller T. If you're not moving fast, the ambient IR shouldn't change enough to warrant more than 50 pps or so. Also, you have a microcontroller. If you want an LED to flash at a certain distance or timing, program it. You shouldn't need to fuss with analog comparators for something like this.

I like what you're doing and wish I had the free time now to do the same. Keep us posted.



I hadn't thought of pulsing the IR LED, thanks for the suggestion! I noticed the TCRT5000 modules use pulldown resistors instead (active high) compared to the regular sized infrared pairs with pullup resistors (active low), but they should work either way.

I had used a variant of the circuit I posted to process the photoresistor voltage divider for the edge detectors for one of my sumobots. I could have just fed the voltage divider directly as analog input but I was lacking pins at the time and needed to make sure the signals were all processed so all the microcontroller had to do was make a decision. I have 8 IR pairs, I don't need all working at the same time so I'll use SPDT switches to not go over the pin count.


File: ea1f5a1785a1bcb⋯.pdf (7.01 MB, P-009-TPC.pdf)

Very useful information I found attached (it's a "real" pdf lol, no need for renaming).

They have two pulses for the diagonal IR, a second power pulse will make it detect whether the mouse can move diagonally to save time. Awesome.

I am seriously contemplating using stepper motors since they are accurate and highly recommended. Competition micromice apparently use the full suite high speed DC motors + encoders + PID. I was originally using DC motors without encoders, just winging it hence the additional side IRs to keep equidistant between two walls. We'll see, a mark II design that fits the larger steppers might be in the works if I find lots of trouble here.



You have 8 pairs of IR LEDs/sensors, and not enough pins to drive/read them, so you want to disable some with a switch? Have you considered a shift register like the good old 595? Only needs 3 GPIO pins and no passive components to drive 8 outputs. Can be bit-banged in a handful of lines of code. You may be able to drive the LEDs directly from the 595 outputs if you don't exceed the total package current limit. Otherwise, go with a darlington transistor array on the low side of your LEDs.

See this page for an intro if you're interested: https://lastminuteengineers.com/74hc595-shift-register-arduino-tutorial/


So there's a realtime IB located at chen2.org. It's sort of like a chat system, but also has some IB features. With their permission, I'm considering starting a /robowaifu/-themed thread there and linking to it here on our banner. This could give those of us who would like to a usable means to chat in realtime together w/o having to use discord.


4 posts and 1 image reply omitted. Click reply to view.




i'll take that as a yea.


this seems kind of comfy, i like it


not bad



anything to quit using discord


The IRC is nice and lonesome as of now. I think a lot of anons have issues with freenode.

File: 73d95608477cad8⋯.jpg (250.17 KB, 1170x2048, 585:1024, CatGrillMeido.jpg)

File: 317be8cefda3294⋯.png (1.14 MB, 1200x900, 4:3, ShhhShe'sRecharging.png)

File: b39b1955e158b88⋯.jpg (93.53 KB, 800x540, 40:27, Motors.jpg)

File: f602e65159e3ee8⋯.jpg (75.06 KB, 920x400, 23:10, hydraulicpneumatic-they lo….jpg)

File: 15a50038137c385⋯.jpg (140.44 KB, 490x270, 49:27, ImagineInADress.jpg)


Hello fellow Anons! Kiwi here to provide basic educational facts about various actuators we can use for gifting artificial avatars of our hearts desire motion!

1. Let's start with a personal favorite, the impractical, inefficient yet oh so fascinating: Heated Twisted Nylon!

What are they? They're nylon threads which have been spun around then annealed to seal in their coils. A heating method causes these threads to then contract or expand.

Good: Why is this a personal favorite? Simply put, it's natures muscle substitute for muscles. To elaborate, this marvelous unvention contracts like human muscles, has a similar practical strength/weight/volume as human muscle. Icing on this proverbial cake comes in its incredibly low cost of manufacture. Materials needed are nylon threads and a heating element. A fixture for production can be produced simply, operated with incredible ease, all while having a low cost. It very well could have revolutionized all of robotics if it weren't for its flaws.

Bad: This is honestly a terrible actuator. Its greatest flaw comes from its speed. they aren't as fast as human muscles unless they're underwater. Water reduces efficiency to unacceptable levels if they're powered by batteries. Water is also rather heavy. If used, you'd have a waifu that moved slowly , would seize up in hot weather, and her battery would die rather quickly. Final nail in the coffin: it's very difficult to get positional control.

2: Pneumatics, moving her booty with air!

What is it? Pressurized air is guided to an actuator where its energy turns into motion. Popular air actuators include rotary turbines, cylinders, and air muscles.

Good: Actuators are light for their power. Positional control isn't difficult to attain. They can be faster then human muscles. Heating elements can be used to augment performance to higher levels.

Bad: These things require electrical actuators to function properly. Thus, they're inherently more complex then electrical counPost too long. Click here to view the full text.

30 posts and 9 image replies omitted. Click reply to view.


File: 482ff770e4c2536⋯.png (579.96 KB, 750x844, 375:422, howtoevencompete.PNG)

File: c48a014f173f8de⋯.gif (2.8 MB, 350x198, 175:99, aerogel waifus be like.gif)

>Aerogel artificial muscles can contract up to 220% their dimensions.


Are you ready to be carried by your robowaifu, anon?

Electroactive muscles are the way!



>By the way, is there a discord server for this topic?

Have a look at this thread and see what you think anon.




>Are you ready to be carried by your robowaifu, anon?

Why yes, yes I am anon. Thanks for the link looks promising. Again, hopefully, in the intervening years since 2011 they have managed to create an economically-viable product we could actually use as hobbyists. Please keep us informed Spain-anon.


File: 0518278922bfc9b⋯.jpg (626.77 KB, 646x1200, 323:600, despacit0.jpg)


Sorry to keep you waiting guys, I've been doing some research, I'll post a resumee of my journey:


So, first thing was to read the whole paper on these aerogel artificial muscles, which as they say, provide a giant stroke, indeed, although they require a high voltage to produce the mentioned strokes. Still, not so bad, so why not buy a bit of this material for test purposes?

Well, you see, even a decade after its discovery it is still quite expensive, since only 25g cost 300$, and 1kg cost 1000$. Here, have a look for yourself:


In my case I discard the buying option, I've seen a bit of its production methods, it doesn't look so complicated to me, so I'm still researching this option making these nanotubes forests on my own, which I'll post updates when I get to a conclusion. Video on how these nanotubes are made:


On the other hand, I've been researching a bit these HASEL actuators for artificial muscles, and I've a few models in mind, I'm researching in detail for materials to build strong artificial muscles inspired in these HASEL actuators, I think I might get artificial muscles as strong as human muscles that will look alike the muscles in the video and good enough for a housewife's deeds.

Post too long. Click here to view the full text.



Meant to say summary instead of resumee

File: 05f262df79ead66⋯.jpg (81.73 KB, 600x900, 2:3, DeMuc7LUwAAtyTB.jpg)

File: 3a794797e428dff⋯.jpg (34.08 KB, 445x640, 89:128, bd4773476c1e884a944861e3a1….jpg)

File: a670f840d9fd776⋯.mp4 (429.16 KB, 484x360, 121:90, x3AVqafG1egHhF8Q.mp4)


ITT, post links and files of either personal or found robotics projects from the internet. I'm collecting resources as inspiration for building a weeb-themed robot head, torso and finally arms.

> Weeb Robots.



> General Humanoids



> Eyes and Hands



I'm basically trying to compile a lot of videos and files relating to many different humanoid robotics projects so I have a basis of where to start. I have a 3D printer but am currently unsure on materials, thus why I'm looking for inspiration.

99 posts and 43 image replies omitted. Click reply to view.




I think we're all onboard with this for the present anon. I would love to have a robowaifu mostly indistinguishable from a beautiful young woman, but that's not going to happen anytime soon, so for now cute irl animu catgrill meidos in tiny mini skirts are our best option–and will be a fun alternative even in the future.




Is she a robowaifu? Care to post some pics of her with context in the 'RWs in Media' bread if so?


File: 11dc3462f8b8ec8⋯.jpg (38.91 KB, 591x522, 197:174, 1520708449670.jpg)


She's the robogirl you play as in Contra: ReBirth, but she's fairly obscure, as little art of her exists.




It's a doll some Japanese guy is building. Nothing electrical, however it apparently has a good range of motion; so theirs the potential to add some actuators.



Thanks anon, some good references there. He is obviously very dedicated and I wouldn't be surprised if he could be convinced to take the robowaifu-pill tbh.

File: b50d09ed6b6aeb8⋯.jpg (485.35 KB, 1053x1400, 1053:1400, Joybot.jpg)


What would be a good RW simulator. I guess I'd like to start with some type of PCG solution that just builds environments to start with and build from there up to characters.

It would be nice if the system wasn't just pre-canned, hard-coded assets and behaviors but was instead a true simulator system. EG, write robotics control software code that can actually calculate mechanics, kinematics, collisions, etc., and have that work correctly inside the basic simulation framework first with an eye to eventually integrating it into IRL RoboWaifu mechatronic systems with little modifications. Sort of like the OpenAI Gym concept but for waifubots.


16 posts and 7 image replies omitted. Click reply to view.



>and swarm the world with Waifu-1000 terminators

Would.. would this really be a bad think anon?



I thought that this might be relevant to this board


It basically means that if you'd give this A.I. a body it'd learn to walk awkwardly on it's own, if it was smart enough we could program our waifu by just gicing feedback like don't swing your arms too much and not to run as fast as possible.



Yes, that's very relevant anon, and thanks for taking the time to check the catalog and put the link into the right thread. If this is an area of interest to you, then you might check out the OP in the bipedal bread as well. There is some simulation-based research that has made some progress with automated bipedal kinematics training.



Isn't that thread more focused on the mechanical side of legs and bipedal motion? I'm kind of more into the teaching side of A.I., what's the best thread for posting these sort of suggeations?



>focused on the mechanical side of legs and bipedal motion?

That's probably a legitimate observation. From my perspective the two (AI & Mechanics) are inextricably intertwined. The OP has images and links to related research.

>what's the best thread for posting these sort of suggeations?

There are about 6 or 7 AI-specific breads on the board last count. >>4658

This bread is good if the goal is a 'virtual training gym'. Using any of the others or creating your own if none of them are specific enough is fine.

File: 7e96c61ff404adb⋯.jpeg (103.12 KB, 700x882, 50:63, prosthetic_leg.jpeg)

File: 3380f929e98f9d8⋯.jpeg (147.49 KB, 333x500, 333:500, kevlar_roboleg.jpeg)

File: 87f2cc6b2aeb83b⋯.png (167.18 KB, 982x664, 491:332, Flexible Muscle-Based Loco….png)

File: 714f57f44d722af⋯.png (97.41 KB, 1062x400, 531:200, SA2013.pdf — Flexible Musc….png)

File: 6053ec98ac088f5⋯.png (195.69 KB, 1024x714, 512:357, SA2013.pdf — Flexible Musc….png)


We need to talk about bipedal locomotion. It's a complicated topic but one that has to be solved if we are ever to have satisfyingly believable /robowaifu/s. There has surely already been a lot of research done on this topic, and we need to start digging and find the info that's out there. There are some projects that have at least partial robolegs solutions working, but none that I know of that look very realistic yet. We probably won't come up with some master-stroke of genius and solve everyone's problems here on 8/robowaifu/, but we should at least take a whack at it who knows?

I personally believe we should be keeping the weight out of the extremities – including the legs – while other anons think that we should add weight to the feet for balance. What's you're ideas anon? How do we control the gait? How do we adjust for different conditions? What if our /robowaifu/ is carrying things? What about the legs during sex?heh Should we focus on the maths behind MIP (Mobile Inverted Pendulum), or is there a different approach that would be more straightforward? A mixture? Maybe we can even do weird stuff like reverse-knee legs that so many animals have. RoboFaunWaifu anyone? What about having something like heelys or bigger wheels in the feet as well?

I'm pretty sure if we just put our heads together and don't stop trying, we'll eventually arrive at least one good general solution to the problem of creating bipedal robot legs.


ITT post good robowaifu legs

>tech diagrams sauce



68 posts and 27 image replies omitted. Click reply to view.


YouTube embed. Click thumbnail to play.

I've been following Robo-one style wrestling robots and plan to make a variant to learn and become proficient at bipedal robots, here's one fully 3d printable (the yahoo link goes directly to stl downloads).

Very useful! One thing I'd do first though is adjust the braces instead of fitting KRS-2552RHV 14kg-cm servos that cost $69 each, I'd use either HJ3315D (15kg-cm) or RDS3115MG (17kg-cm), both each around $11 when I first bought them (though I haven't yet tested them for all you know those servos suck but even Japanese brand servos are all china-made).



That thing is more agile than I expected and I like how it can pick itself back up when it falls. I hope you can try out the cheaper servos with the design anon.


YouTube embed. Click thumbnail to play.

Harvard's Wyss research lab has some interesting exoskeleton stuff tbh.


YouTube embed. Click thumbnail to play.


YouTube embed. Click thumbnail to play.

File: 6033c346084f856⋯.jpeg (2.29 MB, 3456x3456, 1:1, hi_class_hows_everyone_th….jpeg)


In this thread post links to MOOCs, tutorials, forums, and general learning resources about creating robots (particularly humanoid robots), writing AI or other robotics related software, design or art software, electronics, Makerspace training stuff or just about anything that's specifically an educational resource useful for anons learning how to build their own robowaifus.


ITT we mek /robowaifu/ school.

48 posts and 17 image replies omitted. Click reply to view.




God, I can't wait for domestic robo-waifus~


There's an international organization of researchers focused on these topics. They also conduct the "International Conference on Climbing and Walking Robots and Support Technologies for Mobile Machines" events.



File: 0c23344d6141ddf⋯.pdf (3.69 MB, A-Roadmap-for-Cognitive-De….pdf)

A Roadmap for Cognitive Development in Humanoid Robots

2010 Springer-Verlag


Wyss Institute for Biologically Inspired Engineering

Harvard research lab. Located via anon's post in actuators bread.



File: db11d79641c31f1⋯.jpg (409.29 KB, 1136x1700, 284:425, chobits-yeo_008.jpg)


My favorite robowaifu is Chii. I'd like to see yours.

83 posts and 159 image replies omitted. Click reply to view.
Post last edited at



>Honestly if I could just get a robowaifu that can make coffee, that would be an excellent start.



File: e48d514b9d953fb⋯.gif (3.22 MB, 680x383, 680:383, 356e8d51f223ad54275c83cdbf….gif)


There's a game called Subverse that just made over $1.2 Million Dollars on Kickstarter despite being a porn game.' Best part is that one of the main heroine's and a big part of their marketing material is a robot-cutie. Not only is this great because it might fuck with the AAA industry and start a positive trend. But I'm hoping we see a lot more men realize the appeal of artificial waifus. Because more demand for a product can only help us.

Better yet, it just goes to show how we don't need super realistic waifus. A stylized one will work better with current technologies. Making plastics a very viable option for the majority of our waifus bodies. Hell, even doing a BJD style robot with soft TPE sections would work well.



Thanks for the heads up, looks like she's easily the best grill.


File: 5017cd9a3b39244⋯.mp4 (8.17 MB, 1280x720, 16:9, Subverse Cinematic Trailer….mp4)

Definitely, hell I backed the Kickstarter just because of her. That and the trailer is freaking amazing. I'm even considering dropping a large amount for the art book so I can get her concept art. Either way, I really hope more game's have robot waifus in the future. I'd love for it to catch on with more men.



Nice. She's both an Illegal Sexbot & the Ship's AI system. Good combo tbh.


>I'd love for it to catch on with more men.

Visual/Virtual Waifus are certainly an important component in an overall advance of robowaifus with more normalfag men.

File: 52981ee6467af46⋯.png (56.78 KB, 960x720, 4:3, materials.png)


I would define a robowaifu as a doll with robotic features. However there are many different types of dolls (BJD, cloth doll, sex doll, etc). A doll has a skin or surface material, sometimes a filler (cotton), and sometimes internal structure (bones and joints).

Continuing the discussion from >>2831 >>2836 , I want to create a thread to explore the many possible surface materials for a waifu (robo or no). The most important decision is whether to use a hard or soft material.

Hard Materials



>3D Printed Hard Plastic (PLA/ABS)

>Injection Molded Hard Plastic

Soft Materials

>Natural Fabrics (Cotton, Silk, Wool)

>Synthetic Fabrics (Vinyl, Polyester, Nylon)

>Fur/Hair (presumably synthetic, inb4 yiff in hell)

>Silicone or TPE Rubber (TPE is basically a cheaper form of silicone)

I'm strongly biased against the hard materials for comfort reasons. Personally, I have a hard time seeing myself falling in love with something hard, but others on this board talk about using hard materials, so I'm trying to keep an open mind.

My preference is for silicone, but there are four big problems with it. Firstly, it's expensive. Secondly, it impedes modification after the silicone has set. Thirdly, it contributes to the uncanny valley/silicone slut issue. Fourthly, it is heavy, and this weight really constrains the skeleton, posablity, and probably robotics. Because of the weight, silicone dolls have heavy-duty skeletons.

My second choice is therefore fabric, presumably stuffed with cotton. Fabric is super comfy, and has no uncanny valley issue. A non-fuggable fabric doll or robot would have no stigma issue, and could be the start of a productive hobbyist scene with plenty of females. FabriPost too long. Click here to view the full text.

44 posts and 20 image replies omitted. Click reply to view.


>ctrl-f 'carbon-fiber'

>no results

wtf /robowaifu/ materials bread?


Post last edited at


File: 2dfe3dc6c42ad6d⋯.jpg (113.92 KB, 368x591, 368:591, Oscar_Pistorius-2.jpg)


Carbon fiber is very difficult to make and requires pretty heavy duty safety gear even for basic cutting, the cycling fanatic community where they build their own custom bicycle frames specific to their body proportions is a good place to see what lengths you have to go to make it yourself. The only place I can see it being used on a humanoid robot is on the legs as a large spring in a similar way that's seen in high performance prosthetic legs.

For use in a sex doll/robot it's pointless.



Actually, I can see it being used on many places in a robowaifu, specifically an particularly the endoskeleton. But yea probably not going to see a lot of use in the early hobbyist scene of robowaifus b/c of expense, not b/c of difficulty.


YouTube embed. Click thumbnail to play.


Lots of information released to general public:


Main points:

Vinyl: cheap molds, expensive process per piece, limit 1000 a month

Injection Plastics: expensive mold, cheap process per piece, 100K+ a month

No plastic model company has ever attempted 60cm tall dolls before apparently, most plastic models are really brittle.

What this means to us: Injection molding is probably out of our price range (until a huge company much bigger than Realdoll gets into the business).

Somewhat related to mixing and matching materials, I'm reminded of this group who created an arduino Metal Gear Rex robot, took them 3 tries…

mk1: metal frame with cladding, motors fried because too heavy

mk2: 3D printed monocoque chassis, still too heavy

mk3: mix of 3D printed parts, metal frames, and foam parts (lots of weight balancing), this is the one that actually managed to stand up.



Good analysis anon. Weight is everything at this stage.

File: 94782f0cf9b28ac⋯.jpg (995.67 KB, 1280x870, 128:87, 1458511968584.jpg)

File: 20835a0a5c2c255⋯.jpg (289.2 KB, 1024x1446, 512:723, 1482808911923.jpg)


This is a thread to discuss smaller waifu building problems, solutions, proposals and questions that don't warrant a thread. Keep it technical. I'll start.

Liquid battery and cooling in one

Having a single "artificial blood" system for liquid cooling and power storage would eliminate the need for a vulnerable solid state battery, eliminate the need for a separate cooling system, and solve the problem of extending those systems to extremities.

I have heard of flow batteries, you'd just need to use a pair of liquids that's safe enough and not too sensitive to changes in temperature.

This one looks like it fits the bill. The downside is that your waifu would essentially be running on herbicide. (though from what I gather, it's in soluble salt form and thus less dangerous than the usual variety)


How close are we to creating artificial muscles? And what's the second best option?

Muscles are perfect at what they do; they're powerful, compact, efficient, they carry their own weight, they aren't dependent on remote parts of the system, they can be controlled precisely, and they can perform many roles depending on their layout alone.

We could grow actual organic muscles for this purpose already but that's just fucking gross, and you'd need a lot of extra bloat to maintain them.

What we need are strands of whatever that can contract using electrical energy. Piezo does the trick at small scales, but would it be enough to match the real thing? There have been attempts, but nothing concrete so far.

What are some examples of technology that one could currently use instead?

High level andPost too long. Click here to view the full text.

214 posts and 95 image replies omitted. Click reply to view.



>Do you have any specific suggestions atp on planetary gearing you like with one of the small brushless motors?

Sorry, wish I did. I've actually shelved the whole BLDC thing for the time being until I can get better set up to work with mechanical stuff. I need some more tools and a 3D printer for sure.

When I do though, I'd like to try to replicate the opentorque guy's experiments with a bench mounted motor first. Also, I read that somebody somewhere had success with coupling some outrunner BLDC motors to cheap brushed motor gearboxes like you see here: https://www.banggood.com/ZGB37RG-DC-12V-2-1500RPM-6mm-Shaft-Dia-Cylindrical-Permanent-Magnet-Reduction-Gear-Motor-DIY-Robot-p-1370560.html?cur_warehouse=CN but I haven't been able to find any more info yet, nor the specs (input shaft size, etc) or where to buy these little gearboxes themselves. So I'm not sure if those can be used or not even.


File: 08e8cda2a30d7e8⋯.jpg (76.57 KB, 1280x720, 16:9, Shion.jpg)


I've realized this, though I thank you for posting this information as it's important for our goal.


Cost of actuators is right up there with AI development.They're both huge obstacles to the creation of waifus. This is the best I could find after much research. https://www.dx.com/p/high-torque-dc-brushless-gear-motor-black-grey-2073605

I couldn't find any decent reviews online and I'm hesitant about it given its lack of documentation and being a generic Chinese part. I just have had bad experiences with Chinese parts.


3D printing isn't ideal for gear systems, I've tried, due to 3D printing being inherently plagued by inaccuracy, rough surfaces, and an inability to withstand any substantial torque. Because of this, the gears either have to be huge or, you use an expensive printer with expensive filament and go through post print polishing, Sadly, it's most effective to just buy pre-made gearboxes/gearmotors.



>Sadly, it's most effective to just buy pre-made gearboxes/gearmotors.

perhaps that will change sometime in the future. for this stage, sunk costs in r&d are a normal fact of life.


YouTube embed. Click thumbnail to play.

Hexbots have a ton of versatility and charm.

Here’s an open source hexbot for about $250:


I could see adding an upper body and making a sort of drider waifu



>and making a sort of drider waifu

Yea, anon mentioned an idea sort of like this too. Thanks anon.


YouTube embed. Click thumbnail to play.


Robotic control and data systems can be run by very small and inexpensive computers today. Please post info on SBCs & micro-controllers.





47 posts and 21 image replies omitted. Click reply to view.
Post last edited at


File: 512b1148008d969⋯.jpg (254.62 KB, 1024x768, 4:3, sparkfun.jpg)

Sparkfun board with Tensorflow Lite

recommended for beginner voice and gesture recognition




That would be an exciting development if an open-sauce version of a TPU is happening? Is that what's going on here anon can you tell? Also, I wasn't very clear on the translation to English of this page. Is he saying this does or doesn't phone home? Thanks for the information anon.




Nvm. Found the English description. It's not a TPU afaict, and it doesn't rely on Jewgle servers.




Ah sorry about that, I rely on Japanese twitter to vouch for products I'll potentially use… if they mention it, that means it does not suck. It's cheap because it's non-cloud, and I admit I'd rather have pre-trained off-line modules since I want my robowaifu to be a useful box even without internet, like the computers of yore.



no apologies necessary, and please keep us up to date on things like this. as far as cloud connections, etc., go, not wanting botnet in my robowaifu is exactly why i haven't just used the framework by MicroShaft and similar till now to create a chatbotwaifu section.

File: 8dce7ff484fbe95⋯.jpg (20.46 KB, 416x350, 208:175, Chibi-Madoka-Kaname-Doll-P….jpg)

File: d1198f092a12c17⋯.png (7.91 MB, 1835x2581, 1835:2581, sayaka_miki__magical_girl_….png)

File: 48d5f24f41000c3⋯.jpg (37.14 KB, 500x500, 1:1, 9cfd23397a76010c82c687e4ea….jpg)


Thoughts on making a paper waifu then adding robotics? I want animu grills but, most robots have uncanny 3DPD faces that aren't nearly as cute as a real waifu. With paper/screens, atleast the face can keep the purity and beauty of 2D.

93 posts and 78 image replies omitted. Click reply to view.


File: 7a5710494b3e516⋯.jpg (3.75 MB, 3456x4608, 3:4, IMG_20180603_203836.jpg)


There'll be new updates tomorrow, hoping to make good progress.


File: 05d7fbcb9321303⋯.png (113.98 KB, 552x720, 23:30, Imanouti_gumi.png)


Wrong pic




Neat. That looks like it would have pretty good leverage available for the package size. Thanks for the updates anon.


How you doing Papercraft Bro? You still tinkering with your project?


I still think OP's idea is literally the best approach for a beginner robowaifu. It keeps the shell weight down to an absolute minimum, which is important from an engineering aspect tbh.

File: 3c1d6964de2c416⋯.png (317.79 KB, 1024x747, 1024:747, 2000px-Weakness_of_Turing_….png)


What resources are there for decent chatbots? Obviously I doubt there would be anything the Turing Test yet. Especially when it comes to lewd talking. How close do you think we are to getting a real life Cortana? I know a lot of you guys focus on the physical part of robo-waifus, but do any of you have anything to share on the intelligence part of artificial intelligence?



>I know a lot of you guys focus on the physical part of robo-waifus, but do any of you have anything to share on the intelligence part of artificial intelligence?

>A quick check of the catalog reveals;

<AI centric







<voice & language centric




<software centric



<And there are several others with lots of good, related material, for example



Post too long. Click here to view the full text.



>>130 belongs one lower down



Also, there is >>1648 for a general, and there is a Sepples and Python bread. Not really AI per se, but they both are fundamental to the AI industry.


Maybe a stupid question, but what are /robowaifu/'s views on mycroft.ai voice assistant?

From what I've heard, it is open source, self-hostable and customisable.



update: found the thread for it in catalog

File: 838a8c9ddce6cf2⋯.jpg (144.08 KB, 1300x1035, 260:207, robowubs_plushie.jpg)


Will it be possible in the future for AI to become sufficiently advanced to feel real emotions? We could probably simulate a reasonable approximation even now to be a gratifying enough substitute for her master in their relationship together, but hypothetically speaking, could it ever turn into something real as an experience for the waifubot herself?


>Robowaifu: "I love you Oniichan!"

>Anon: "I love you too Mikuchan."

true or false?

55 posts and 38 image replies omitted. Click reply to view.


File: 70e41733658c001⋯.jpg (138.71 KB, 1440x1080, 4:3, 70e41733658c001f3315500823….jpg)


>a year late

Almost to the day, too. That's alright though, I'm still here.

Affection could be considered a side-effect of friendship, but it's still different enough to warrant a seperate classification. Plus, it's something that different people will have different opinions on what this should look like, to which I bring back my argument of a modular system.

Loyalty, again though possibly a side-effect, is still different enough to require seperate classification. Unlike other components, however, the concept of loyalty is unwavering and static, and as such should be made so in programming. Devotion, though perhaps a stronger version of loyalty, is similar to affection in that people will have wildly different opinions on how this should look, and how strong it should be. With this in mind, it should be classified seperately.

For common intrests and beliefs, however, I will concede that there may be better and more efficient ways of going about it. Looking back, they would have to be so tailored that each owner would have to do the coding himself. Therefore, I now think that It'd be best to have a learning system instead of modules, so that the owner could teach the AI about his interests, similar to how he would with a real person. This will also have the added benefit of providing an organic bonding experience.


You're only painting yourself as an antagonist by introducing yourself in a manner so hostile, and you kill your argument with ad-hominems and unfounded theories.

You speak of how his sources are tainted by modern acedemia, which you claim is inherintly corrupt. You speak of the government keeping technological secrets from us, and how the state is performing secret tests and making developments to further their own power, and subdue us. I will not argue against these points, as I wholly believe them, same as everyone else on this board (to varying extents).

However, you link back to occult studies and cult beliefs on how the world really works. Have you onPost too long. Click here to view the full text.



A learning system would definitely be a very attractive feature. However, I don't think it is an absolutely essential one. Surely it's something that people are working on regardless, so I'm not terribly concerned. However, if available, then yes, it would obviously be something many people would want their robowaifus to have.



>Can Robowaifus experience love

No, but neither can real women, so who cares.


File: cd031ea3a91ba4f⋯.gif (389.21 KB, 450x343, 450:343, She'll be here forever.gif)


I would posit that it is necessary for any advanced AI to be capable of feeling love,and furthermore to feel said love for at the very least a subset of humanity.

Such is the only solution to the issues created for us by bringing such existances to life.



I think I get the point you're making anon. I'm just not sure real life actually works that way.

Delete Post [ ]
Previous [1] [2] [3] [4] [5] [6] [7]
| Catalog | Nerve Center | Cancer
[ / / / / / / / / / / / / / ] [ dir / animu / baaa / choroy / dempart / monarchy / tingles / vg / vichan ]