This 7 axis Robot Arm finally works!!! (part 3, final?)

[@Skyentific]

Like it, share and subscribe! :)

[@WillianMai]

why you didn't use the pulley reducer from the video kzbin.info/www/bejne/q6Wnkpqlqt2Aa7c "You've never seen the Robot Joint like this one!"? are there any problems with that design that led you not to use it? I am asking you because that design is my favorite, but you are the one that tried it and know if it really works with no backlash

[@bertbrecht7540]

Thank you for showing when you make an error. 90% of my time is spent on debugging and most of the time I would classify my errors as simple. Most teachers don't want to show their errors but it is where I learn a lot.

[@Skyentific]

I have the same. Sometimes I stack for couple days over one simple error!

[@СергейПаюнен]

Вери гуд джоб май френд))))

[@patrickderp1044]

i saw the brushless planetary video and in the comments someone linked the cheetah.....and here you are implementing it! how awesome

[@klausnielsen1537]

I knew you would make a great comeback. Well done Mr. Skyentific. Love your projects and all your ideas. Thank you for sharing.

[@thiagoennes]

The robot ACTUALLY did the robot ! :D

[@haroldfinz4863]

A big well-deerved round of applause... Congratulations, and enjoy the moment.

[@PhG1961]

Very impressive ! Excellent work (in progress)

[@wobblyjellyman69]

Awesome to see this finally come together

[@ahmedwaly9073]

Wooow that's so cool, amazing job

[@lazaroybarra5998]

I am happy for you! You are doing it! Progress

[@nathanthomson7273]

Love the channel! I'm working on 3d printed generator motors, compound gearboxes are next. Keep up the great work.

[@tonysparc5322]

Learning english with youк videos ))) Thank you very match!

[@sabtvg]

I think vibration is related with the mass of the arm. You want to decide a constant speed but the speed should change according to arm mass. Accelerated and deaccelerated to avoid intercity. It means you should previously know the distance you want to move axis 1 so you can planify the speed curve before you start the movement. I saw many of your videos, thank you very much.

[@keetelemetry]

another milestone. great job

[@Hackwave]

Looks awesome! great work

[@keepfighting8402]

This is amazing! Great work!

[@vincentsoubbotin7830]

inverse kinematic calculations with all of those offset centers of rotation are going to be hell, have fun

[@Wahweka]

Just solve it numerically?

[@vincentsoubbotin7830]

@@Wahweka That's true, I was thinking of analytical solutions, however I think even numerical solving would be eased if the adjacent axes were coplanar to each other

[@OhHeyTrevorFlowers]

Woot! Great progress.

[@stefanguiton]

Great work!

[@Skyentific]

Thank you!

[@PiefacePete46]

Well... you would have to call that a success! Many thanks.

[@JamesNewton]

Nice! May I suggest setting up a counterbalanced arm like the Haddington Dynamics Dexter? The idea is to put all the motors on the /opposite/ side of the joints they turn so that the arm is counterbalanced. The same belt tricks you've used with the end arm, can be applied to the major joints combined with moving the actuating motor on the opposite side of the joint. On Dexter, we use cone drives for Joints 1 and 2, then move the motors for joints 3, 4, and 5 to the opposite side of joint 2, but there is no reason you can't do that for each joint: Move the motor for Joint 2 to the opposite side of joint 2, Joint 3 motor opposite joint 3, and so on.

[@HappyMathDad]

I really like your videos. You keep it informative and do cool projects. Great content for kids too. You would be surprised to know how young some of your viewers are.

[@rayeaglenz]

Well done, nice progress, always interesting, thank you 😊

[@kirilchi]

Good progress!

[@World_Theory]

I too like the idea of the arm with heavier motors towards the base of the arm, with progressively lighter ones towards the extremity. (Assuming weight = power) It reminds me of how the staging of multistage rockets are set up. Could you adapt a rocket equation to robot arms to formalize it? Using, distances, power, leverage, etc.? Evolution seems to generally agree that thick to thin is optimal as well. When was the last time you saw an animal with huge forearm and puny shoulder? (Feet don't count, because they have other purposes, like stopping the animal from sinking into soft ground, which is rather difficult with needle-point limbs when you're not the size of an insect.)

[@hatemzaatra5894]

Very nice

[@i-make-robots]

Congratulations!

[@jonasaskenobis]

Great video, thanks :)

[@sebastiandelavega4978]

great great work.... !!!!

[@Skyentific]

Thank you!

[@Dagor]

I’d like to see a video of the inverse kinematics!

[@ahmedyasser3684]

great Design

[@Skyentific]

Thank you!

[@rickalan3467]

I really like this

[@kayaonur5657]

Hi Gennady, thanks for the video! I think the bulkier the axes get, the more we need to pay attention to the structural elasticity of the joints especially for the axis 1 and 2.. and perhaps tune the PID parameters accordingly. Have you thought about that? Like do you know a basic tool to make elastic calculations to find eigenmodes, bending modes of the structure, like beam and concentrated mass analyses for multi axis arms?

[@Skyentific]

This is all very complicated for me. I need to learn more on this.

[@kayaonur5657]

@@Skyentific if I ever get to your level, I will learn and make a video. But for now I think ROS and inverse kinematics are lower hanging apples :) Thanks again!

[@Chris-bg8mk]

Inverse kinematics next?

[@HerrPOZA]

Yes, please!

[@Captaincookie3]

No, please continue with this design. It's a really cheap alternative for not so precise industrial applications and other non-industrial uses. Atleast it's what I think. Considering the sky rocketing prices of industrial robot arms.

[@jcugnoni]

Congrats. This is great ! Maybe you could try to implement a ROS interface ?

[@MaximDybarskiy]

Could you please design a robotic gripper? There are some cool adaptive grippers like RH-P12-RN. You can move it to the next level.

[@kstansimit4252]

can try to check payload and accuracy/repetitive of this arm?

[@gatoninja4387]

give him a test of strength

[@Skyentific]

Yes, I will!!! :)

[@gatoninja4387]

@@Skyentific Are you planning to tackle the problem of robotic hands? or you will solve it with the typical tweezers''''??

[@SimonBirrell]

Great videos. Is this still your recommended path for obtaining a relatively low cost torque controlled arm? I liked the look of the Amber B1 but the price is now around $11k, putting it in the range of the Panda Emika. I don’t need the world’s best arm, but something below $5k that can do real, light-payload tasks.

[@MichaelDeeringMHC]

Instead of turning little knobs, you should make a frame you can wear to control the arm.

[@bertbrecht7540]

Is there already a multi axis arm that is powered by ONE motor and no actuator? Where the torque is split up (perhaps via clutches) to each axis? This could be a nice way to keep weight low in the mid and outer extremities of the arm.

[@Skyentific]

Many people had this idea. But I never saw a real robot arm like this. I think it would be complicated to design such arm.

[@killerdoxen]

There was a toy around the ‘80s from Radio Shack called the Armatron that had a drive system like that. The 8 Bit Guy here did a repair of one of those. You do see how it was put together.

[@kayaonur5657]

if there was a precise and speed-adjustable hydraulic actuator, wouldn't that be awesome :)

[@Skyentific]

And inexpensive. :)

[@tvathome562]

very cool, how much is the cost of build please

[@DMonZ1988]

ah man i was so excited to get this notification. what about the design you spent so much time working on? this arm just seems like a distraction from the far better one you were designing before :-(

[@Skyentific]

For this design I spent really a lot of time. And I was extremely disappointed that the video performed really badly. For some reason people was not watching it. Bit I had 5 axis working. I was expecting at least 100 thousands views, but I got only 30 thousands. I know it still sounds a lot, but I think it is not enough for the amount of work I put in. So I will just forget this bad experience and continue with this design, which you liked. :)

[@Captaincookie3]

@@Skyentific No, please continue with this design. It's a really cheap alternative for not so precise industrial applications and other non-industrial uses. Atleast it's what I think. Considering the sky rocketing prices of industrial robot arms.

[@alfiolocatelli5093]

Hello, great video, will you implement inverse kinematic like the other robotic arm you made? It would be amazing if you can provide a video where you explain in detail how you create and write the code

[@mohitjoc25]

Subscribed

[@noithugian494]

You can make videos about robotics course?

[@erikthompson404]

One comment of support.

[@tonacho77elcapo]

Excellent video! How do you know which belt you should use and if it will handle the torque? Thanks!

[@rkavinkailash4587]

Integration with ROS (MoveIt) next?

[@Larock-wu1uu]

Amazing project! Too bad it's so bulky and slow. The great advantage of brushless motors is their power density, which should allow for a lightweight and agile robot arm. Maybe in a next project..? :)

[@naij8124]

very cool design! will you be making the arm trainable and log encoder values?

[@ДмитрийПантелеев-д9я]

Работа внушает уважение! А что за музыка играет, не подскажете?

[@Bananananamann]

Dear Skyentific, I really like your Videos, your channel is an inspiration and also quite unique in the interwebs. Can you recommend (a) any specific rotary encoder (b) a source for axial and radial bearings and (c) a student friendly dc-motor-rotary-encoder-combo or servo motor?

[@nguyenminhthanh1122]

can you share the mane of motor driver?

[@saveplanet3977]

Do you also impart any kind of training sir I can enrol to

[@orlandojam8]

how fast can it move?

[@ericlotze7724]

Obligatory Comment to Feed the KZbin Bot's Bloodlust

[@qborki]

I see you've moved the motors closer to the base of the robot to decrease the weight. Why not go further and have ALL the motors at the base. You could transmit motion with some tendon-like wires thus making the arm even lighter.

[@Skyentific]

This is a great idea, but it is very difficult to accomplish. There are very few robot arm, which do this, mainly because the design becomes complicated. Also when you have long tendon-like wires, the stiffness of the arm decreases, as even steel wires stretch. I know the WAM robot arm and LIMS2 (ambidex), which use this method.

[@alexzuzan]

Крутяк!

[@gordonpromish9218]

it looks to me like you will inevitably run into the limits of the mechanical properties of 3D printed laminar plastics. though it is convenient for rapid prototyping of shapes, it is junk compared to machined metals when making load-bearing parts, requiring excessive bulk to (poorly) handle the working forces.

[@Skyentific]

I think you right. But I still have some ideas to test!

[@gordonpromish9218]

@@Skyentific - good luck!

[@rocketraccoon9596]

What’s ur 3D printer

[@DPTech_workroom]

Американцы не поймут скотч... 🙃 Или они так же говорят? Double sided tape

[@Skyentific]

Они понимают. Но тэйп правильнее ;)

[@WillianMai]

why you didn't use the pulley reducer from the video kzbin.info/www/bejne/q6Wnkpqlqt2Aa7c "You've never seen the Robot Joint like this one!"? are there any problems with that design that led you not to use it? I am asking you because that design is my favorite, but you are the one that tried it and know if it really works with no backlash

[@joe_limon]

Great job what is the build cost?

[@Skyentific]

A lot. Each actuator is around 300-400USD, and there are 7 of them. Plus 2 slew bearings (100usd each). So total is around 2’500 USD.

[@TS-kg4lf]

Still a good "socks" programming 🤣

[@timmyg37]

I wonder who said let's get started 1st. You or makers muse

[@TS-kg4lf]

BTW make two of them for a cooking robot dual arm 🥞🍕🌭

[@selivermedia]

👍👍🔥👍👍👍🔥🔥🔥🔥

[@shuaization]

Слава роботам!!!

[@guardianpresencerobotics7776]

just make the wrist move a little bit better cover up the gear drives belts water proofing and start making tools changes for the wrist i would like a string trimmer for the wrist .

[@elucky51]

I like what you do, it's very interesting. But what I like most are your socks.

[@gerokatseros]

Well , you started building an affordable robotic arm, and you ended up putting 500$ actuators for every join!!! This is called sponsored my friend , but if u are to spend thousands of dollard on those components why 3d print? I m sorry it went this way

[@swedensy]

A big misunderstanding among robotics. We live in 3 dimensional space, claiming to have 4 or 7 axis is a lie.

[@samjanssens1509]

so its a dab-robot ?

[@MJCNCDIY]

oke

[@詹勳昇]

can I buy this 7 axis Robot Arm ?

[@Skyentific]

It is going to be expensive, as I have only one. But for 10’000USD I will agree! ;)

[@dangerous_ngga]

@@Skyentific what is expensive ?? 4 motors kinda like 500$ so 3d printed arms ~ 100$ )

[@rodstartube]

which motor you see the best?

[@rodstartube]

hope

[@ericlotze7724]

@Skyentific I'm not certain if you've seen this yet, but i found a dissasembly of one of the actuators you used. kzbin.info/www/bejne/g5nbq2V4n5eFjqM

[@VideoStefan17]

a comment ;)

[@LongPigg]

Algorithm.

[@gatoninja4387]

tosco fragile ,,bulky too based on the hideous kuka robot

[@rodstartube]

1ST

[@Skyentific]

Yes! :)

[@gleblin3164]

You are from Russia ? You just don't have an English accent

[@serhiiserhio9382]

Переходи на русский, а то так и зачахнеш с тремя подписчиками.

[@shuaization]

Не вздумай! Такого внятного английского не найти больше)))