My Advanced Realistic Humanoid Robots Project

[SiliconWizard]

Reminds me of the saying "the body is the TempleOS of the ...."  Oh, never mind, scratch that thought.

TempleOS was actually somewhat impressive in the author's own lifetime though. 

The author was a complete weirdo, but TempleOS is an impressive piece of work in itself that actually "works". If the OP here manages the same level of achievement with his ideas, I'll be impressed too. I just have some doubts.

[Circlotron]

I'm a bit rusty on my Old Testament, but isn't #3 a bit of a show stopper?...

"Thou shalt not make unto thee any graven image"
https://en.wikipedia.org/wiki/Thou_shalt_not_make_unto_thee_any_graven_image

... and then even naming it "Adam" ... really?

Seems to me that the OP is inconsistent with his approach.

The direction was not to make a representation of something for the purpose of bowing down to it and worshipping it. And the first man Adam is described as being responsible for the crap state society has degenerated into. Not a good man by any stretch of the imagination.

[ElectronSurf]

Hi, I think your idea is ambitious, but there's nothing wrong with that. I actually like it when people have ideas beyond possibilities.

I see a lot of nonsense in this topic, like "you should not make it like human"? dude go mind your own business, I'll build my robot like a damn alien if I want to!

So I suggest that you ignore this topic completely, you don't need to explain yourself to the public and trolls.

Go publish your progress on youtube and if you had any electronics related questions, part them in chunks and ask in stackexchange to get some decent answers.

Hope this helps.

[artbyrobot]

I just came up with a cool alternative way to downgear a 2430 BLDC motor that might work.

Here's a illustration of the cheap downgearing idea:


So basically, I figured what if I could remove the N20 motor from its gearbox/"gear set" by cutting it free or w/e. But I keep its center axle in place cutting away only everything else. You'd then presumably have a metal shaft as a entrance to the gearbox and a metal shaft exiting the top of the gearbox. I then turn that metal input shaft and output shaft into pulleys. I feed my 2430 motor output shaft pulley/winch into the input shaft of each of 4 N20 motor gearboxes, evenly distributing the load. Each gearbox downgears my 2430 motor 150:1. Each gearbox chatgpt said could handle about 5-6lb load but this can't be sudden or fast direction change this is really pushing it. But it seems 4 gearboxes should handle most of what we'd want from a 2430 motor. And the fact we can fit them all within the height of the motor output shaft default length and within the width of the 2430 motor diameter for the most part seems it would be a pretty significant downgearing for very low space taken as the cost. You could even locate a few more gearboxes off the motor anywhere and have those fed further distributing to them the load if only 4 gearboxes was not enough to handle expected forces. The cool thing is supposing we did this, it would cost us four N20 motors which is $0.80x4 = $3.20. That is VERY cheap for a gearbox as I read that a planetary gearbox for it would be like $25-30! And the planetary gearbox would take up WAY WAY WAY WAY more space which is highly coveted in our application - space we can't afford to spare. And the great thing is these little gearboxes you can fit ANYWHERE into a nook or cranny since they are so tiny... and you can use as many as you want to get up to the total forces you need them to handle as a collective. Seems like this could be a cool technique. I want to give it a go. Any thoughts?

Note: this would be something I'd try on the Dinah robot where I'm using metal gears despite the noise these create since its a lower budget simpler robot I'm doing just to get something done faster for a change. My Adam, Eve, and Abel robots will be going pulleys to downgear to make them very quiet in operation as has been the plan forever.

[Korchahin]

Hello. I have been passionate about robotics and working on AI development for about 20 years. The idea of creating AI and robots is connected to a desperate situation regarding my health. I have a progressive (currently incurable) eye disease called "Retinitis Pigmentosa." After studying existing technologies in gene therapy and stem cells, I realized they hold no prospects for me. Moreover, I have already passed the point of no return. To try and accelerate the process of solving hereditary issues and many other problems, I started researching in this field long ago. I apologize for my poor English; I am from Ukraine, but for some well-known reasons, I moved to Germany. My vision is now very limited, so I left the field of robotics and focused solely on AI programming. As a distraction, my hobby is analog electronics, and I've "caught" the "Perfect Volt" disease I've been reading this forum for a while now.

I would like to share some thoughts on your topic.

Your approach reminds me of someone fascinated by "warm tube amplifiers" in the modern era of 2nm technologies and ultra-high-quality microchip solutions. Take a look at TeslaAI and Figure's developments. You are hopelessly behind industrial giants, and in the end, you won’t achieve anything better and will only become disappointed and abandon this pursuit after a while.

If you really want to build a humanoid, you need to move forward. You're stuck, trying to achieve results where none exist. You’ve drifted too far from the goal.

Move forward with small projects. Just build a simple 5/6 DOF manipulator and learn to control it using a webcam as an eye. Use existing solutions; it will speed up the debugging process. Always keep your goal in mind. Yes, servo motors make noise. But at the start, this is not an issue, and even a noisy manipulator capable of performing some tasks under AI control would already put you ahead of others. Right now, the bottleneck isn’t mechatronics, it’s software control. By achieving intelligent control over the manipulator, you will outshine everyone!

Now, regarding mechanics. If you want to be an innovator, stop using rotating motors. Switch to linear or membrane actuator systems.

Try starting with using the YOLO neural network to recognize objects and place them into a box. Even by implementing such a project, you will understand the features and limitations of the system for future stages. I would be happy to participate in such a project and even wanted to start my own, but due to my poor vision, 3D modeling and mechanical design take a lot of time for me.

etc. Ruslan Korchahin

[SteveThackery]

Right now, the bottleneck isn’t mechatronics, it’s software control.

I've made that argument to @artbyrobot myself, but he seems set on fiddling about with pulleys and string.  He seems to be relying on software which doesn't exist yet, but will be written by an AI. He will write the initial AI himself, and then instruct that to write the rest of the software.

It seems like most of this project exists in the OP's head or in the extensive documentation he has written, along with some pictures he has created in a CAD program.  I think, after ten years, it's time for @artbyrobot to start prototyping some real subsystems - hardware and software.  Even though I have doubts about the feasibility of the project, I would absolutely love to see it work.

[xrunner]

So being directly on topic, let's compare the look of the craftsmanship of current robotic hands already in existence to artyrobot's hand. There are many quality robot hands out there now that have different underlying designs, but I wanted to compare the hand of the robot which is the subject of this thread to a robotic hand of similar design. Since I don't have access to any robotic hands, let's examine which one we might think has more thought, time, and craftsmanship put into it. Although it's subjective, it will give us a good starting point when we look for quality. In other words "I'll know it when I see it" will apply for starters.

Here's a hand that has garnered some attention, although there are many others out there too - let's use this one as an example. On the left is a picture of a robotic hand developed at the University of Washington's Computer Science & Engineering Dept (UW CSE).

UW CSE Robotic Hand

From my point of view, coming from experience in engineering work in my career, it simply screams of quality work and engineering.

On the right side, we have the Arty robot hand, associated with the robot project being discussed here. Both are similar in basic functional design. Does the arty robot hand look different in some way to the UW CSE hand? Does it appear to be a quality piece of work? Which one would you give funding to for further development if you were asked to choose, given no other information? Which one looks like quality work - "I'll know it when I see it" ?

How about you readers out there? I suspect there are viewers out there that might want to share your opinions of what we are following here. I ask you (join the forum if you can) and tell us - which one appears to you to have had the most thought and care put into the design of a robotic hand?

[MK14]

I ask you (join the forum if you can) and tell us - which one appears to you to have had the most thought and care put into the design of a robotic hand?

A sole persons first (or early) robot design or something, might appear to be worse, than something done by people with much greater experience of robot design, considerably higher budgets, and could involve a large team of (well funded) team members.

But, as their (the OP's) experience level builds up, they can make better and better ones, in the future.  But they also might be pleased with their experiences, and later decide to buy a ready made robot, or move onto other things.

It is not that uncommon (I suspect), for inexperienced people, making their first (or early) projects, to be over-confident, in what they can achieve.  How much it will really cost, how long it will take, how much room is needed for the developments and many other factors, that are learnt through experience.

[xrunner]

It is not that uncommon (I suspect), for inexperienced people, making their first (or early) projects, to be over-confident, in what they can achieve.  How much it will really cost, how long it will take, how much room is needed for the developments and many other factors, that are learnt through experience.

He can do whatever he wants and whatever he has fun with to pass the time, as long as it's legal I suppose. We all do that. But like Korchahin said better than me, IMHO, I don't think it's realistic to think he'll have any sort of outcome that will be anything close to what he claims he wants to do. That's my point all along. What he claims to want to achieve in the end is unrealistic, IMHO.

[MK14]

He can do whatever he wants and whatever he has fun with to pass the time, as long as it's legal I suppose. We all do that. But like Korchahin said better than me, IMHO, I don't think it's realistic to think he'll have any sort of outcome that will be anything close to what he claims he wants to do. That's my point all along. What he claims to want to achieve in the end is unrealistic, IMHO.

I agree with you.

Maybe a person, attempting (an over-ambitious) project, is a necessary step, in them coming to their own realization.  That they need to either considerably downgrade the project scope (e.g. buy a pre-made robot, and just create software and/or modify existing software systems, for its vision system, using existing LLM (or what seems best) technologies, to carry out some task or something) or other changes, to make it doable in a practical time scale.

[KerimF]

I wonder if the name Adam of the new robot refers to the well-known story or not.
If it does, Adam, being the first man (as in the story), existed on earth before the existence of the cave men. In this case, the new robot will be another precious piece to be added in museums.

[artbyrobot]

The Dinah robot is coming along well. I modeled the full steel skeleton in CAD to match the dimensions of the Dinah base mesh and created a human bones variation as well to compare that to the steel simplified skeleton and make sure all the joint pivots matched the locations of the human skeleton joints pivot points. With this CAD, I will be able to modify the proportions of the steel skeleton I have on hand. I also added several key additional joints using reference photos of a skeleton I found online. For example I now have 2 pivot points for the knee joint instead of one which gives more clearance when knee bends back. I also gave a few more degrees of freedom to the neck and shoulder area. Here's the CAD:


Note also that I am well on my way to finishing up printing ABS solid infill fingers and wrist bones which I will retrofit onto the steel skeleton so that I can have full 27 degrees of freedom robot hands and wrists to match perfectly the dexterity of the human hand, which is a must.

I have decided that once I finish the arm and head, I will not go on to complete the building of the rest of the robot's body but instead will switch my focus to the AI entirely from there forward. I will code the AI to cause that arm and head to build the rest of its own body.

[RJSV]

   OP, it appears, doesn't live in a sea of doubts and imposed limits;   actually refreshing to read and ponder.   (I trust the author isn't ignorant but simply works while having a confident
style!

As long as problems aren't ignored,  I can admire that style,  experimental but not excessive caution.
)

[SteveThackery]

Out of interest, why keep the ribcage? Surely this is one area where you could improve upon the human original, as it doesn't need to breathe.

[artbyrobot]

I'm not keeping the ribcage or I don't think I will not sure.  The highlighted skeleton is the steel skeleton for Dinah.  The gray skeleton was just a reference point for scaling the highlighted skeleton perfectly to match the Dinah basemesh.  It is not being used in Dinah at all.

[artbyrobot]

Also, regardless of whether I use a ribcage, the robot MUST breath.  It is countless electronics trapped in a rubber suit.  Breathing is the only way to allow expelling of hot air and intake of cool air.  I designed artificial lungs for this reason.  The ribs will not be involved in the breathing actuation though as they are for the human lungs.

[SteveThackery]

Also, regardless of whether I use a ribcage, the robot MUST breath.  It is countless electronics trapped in a rubber suit.  Breathing is the only way to allow expelling of hot air and intake of cool air.  I designed artificial lungs for this reason.  The ribs will not be involved in the breathing actuation though as they are for the human lungs.

Yeah, I was thinking about the cooling, and I think a continuous flow might be better than in-out.  I don't know,  just thinking out loud, really.

[artbyrobot]

Actually, it can't be continuous and here's why:  supposing you are using the mouth/nose to breath, if it continually is sucking in air and simultaneously blowing out air, it will recirculate the same air.  So when it expels the old hot air, it needs a space of time for that to sufficiently leave the immediate vicinity of the nose/mouth area so that the next breath intake sucks in room temperature air instead of the same air it just tried to get rid of.

[SteveThackery]

I was thinking of discharging the air elsewhere. Perhaps through slots or grills with a sufficiently large surface area to keep the air speed nice and slow so the clothes don't flutter.

[artbyrobot]

the clothes would block good airflow and trap heat, very hot air, against the skin reintroducing it to the insides through convection.  Better to blow it forcefully well away from the robot.  Also, by expelling air through the nose, watching it operate outside in winter will be SUPER cool.  Seeing it breath out visible heated air like we do on a winter day will make it seem so alive.

[Andy Chee]

Actually, it can't be continuous and here's why: 

Actually it could be continuous and here's why:

Think of the heart function.  The heart valves permit blood flow in one direction only.

There's no reason why your cooling system needs to be restricted to the same human anatomical bidirectional airway.
 
Indeed, a robot doesn't need a digestive tract, so the entire esophageal system can be redesigned.  As you should be aware, humans breath and eat down the same throat.  That limitation can be eliminated in a robot, and have an air intake through the nose and outlet through the mouth (or vice versa, or indeed reversible).

[MK14]

Seeing it breath out visible heated air like we do on a winter day will make it seem so alive.

I don't think that would work.  My understanding, is that it is the significantly increased water vapor, exhaled through the (damp) lungs, which makes the hot air visible on a cold day, rather than the hot air itself.

I.e. (Robot) Dry air, would be relatively colourless, as I understand it.

[xrunner]

There's no reason why your cooling system needs to be restricted to the same human anatomical bidirectional airway.
 
Indeed, a robot doesn't need a digestive tract, so the entire esophageal system can be redesigned.  As you should be aware, humans breath and eat down the same throat.  That limitation can be eliminated in a robot, and have an air intake through the nose and outlet through the mouth (or vice versa, or indeed reversible).

A robot doesn't need a ribcage
A robot doesn't need clothes
A robot doesn't need to breathe like a real human being
A robot doesn't need a whole lot of useless things the OP hangs on to
All these useless things will keep his ideas for a robot from being realized in the end

[themadhippy]

  Better to blow it forcefully well away from the robot.

theres a suitable orifice on the body thats used to expel waste gases forcefully,maybe you could incorporate one,or is that going to be used  for the  charging socket

[artbyrobot]

As to making it continuous by making separate tubes, you still only have the nose and mouth as options for intake and exhale with those two tubes and if the inlet is always sucking and outlet is always exhaling, theoutlet would send the air right back into the inlet.  The exit holes of the face are too close for this continuous style.  The only way would be to exit or inhale through the ears but I don't like that approach at all.  firstly, it would for example ruin its ability to wear a headset and still breath or wear earbuds to handsfree use a iphone etc.  it's not a good design long term. 


Yes, you would see the hot air mist on a cold day.  Why would human exhalation be "wetter"?  Also mind you the lungs in my design are evaporative air conditioners intermingling with water so that will introduce moisture.

A robot having a ribcage gives a good attachment point for more mechanical advantage for abdominal muscles, rotation of torso, and protects the main computer inside it.  also acts as a good mount for motors and stuff.  So it's good to add.

A robot that is realisitc humanoid passing for human 100% needs clothes lol.


As to things I hold onto, they aren't useless I can prove how each thing is VERY useful to my goals.