Inverse kinematics of a ball balancing robot.

[@kersgames6382]

This is what every bachelor robotics needs for his or her thesis. This was a pain to get my head around at the time. Thank you so much!

[@HandlerNewton]

The vidro is straightforward, simple and smooth, very well edited and the calculations are accurate. Also I love cats. New subscriber.

[@5eurosenelsuelo]

Great video. It's nice being able to see the details.

[@webanon]

Twenty eight years ago I was an engineering student. I graduated and became a licensed engineer. I've always wished I "learned" math and not just the mechanics of math. A project like this would have aided my learning, knowledge retention, and abilities as an engineer. 🍻

[@iosmc4]

Awesome work! Can you talk through your control algorithm next! You’re the man

[@mirkok08856]

Thanks for the video. I always wanted to build such a robot on my own. No as I see the complicated math, I don’t know whether I would have succeeded. Maybe I build yours someday. I’m really looking forward seeing the ball bouncing.

[@amr.a-m8350]

Thank you for video but at 8:20 the functions of [X,Y,Z] than before period at x,z plane when Y=0 , are different and how are derived and the angle theta not clear to get the rotation angle of two motors we would use as function of [d,f] or [X,Y,Z, L3].

[@amr.a-m8350]

?

[@obito_m315]

Hello, this project seems very well finished and takes into account all the scientific aspects of robotics (invers kine, dynamic servoing... excrt). Could you make a small simple video dedicated to multithreading for a raspi4?

[@Marcus001]

This is why the internet is a good thing

[@BryanHaskin]

The AI voice is top notch. It took me awhile to notice something was off

[@pauldolton9118]

wow lol i hadnt noticed where does it sound off?

[@hydropage2855]

@@pauldolton91183:30 for example when it tries to say 0

[@pauldolton9118]

@@hydropage2855 lol i thought maybe English wasnt his first language, nice spot!

[@marjanjovanovic6643]

Thx, very nice tutorial. New subscriber

[@KavinPalanichamy-nj6bz]

Great video. Can you please upload the derivations to Github as well.

[@adib6121]

hello, would you mind explaining how you determined the coordinates of the other two joints? what plane did you use to solve them. would you mind walking me through?

[@Ottomane__cCc]

Have you alternativ servo for rs304md

[@amr.a-m8350]

Thank you for video I have shown again but at 5:22 & 8:20 I need the script or tutorial to get the derivation of the coordinate for the two other motors either the other equations for the two other motors.if there are a code Matlab script for these equations.

[@Koshiro_Robot_Creator]

Thank you for checking. I don't have a MATLAB script, but there is a Python version available on the GitHub link.

[@amr.a-m8350]

?

[@bishoyemadzaka1086]

I need to know the platform circle dimentions

[@eliferin6974]

At 3:40 you say that the center of the platform is in [0,0,h]. How do you know that the center will lie in the z axis? Is this an approximation for small tilt values of the platform or is it always true?

[@Koshiro_Robot_Creator]

The center of the platform is not determined passively; it is specified by us, the users. In this case, we have set the center of the platform to be at a height of h along the z-axis. Based on this setting, we are calculating the motor rotation angles necessary to maintain the platform in that position.

[@iosmc4]

Could you also post calculations?

[@Yousif_Maksabo]

Hi there, Thanks for the tutorial, Is there any book or source to recommend to learn inverse kinematics in general?

[@PetrVacek]

Thanks for sharing. I have the exact setup in my artwork Fluidum (just 85x and synced all together). Can you tell whether it's possible to place platform's center point above the platform't plane? Is it mathematically possible or not? I am not sure.. Thanks.

[@Koshiro_Robot_Creator]

If you could show us what kind of artwork you are referring to with photos or other materials, we might be able to give you a more accurate answer. When you mention the platform's center point being above the platform plane, do you mean that the platform has a very thick underside which is considered the platform plane, and that the tilt and height are controlled by a 3RRS parallel manipulator, with the top center being the platform's center? Are you aiming to achieve a ball balancing mechanism?

[@alrredford1133]

Hi, I have recently created something similar, this video has been very helpful in understanding the maths behind how it works, much appreciated. My only comment was that at 5:26 you showed the maths for one of the other arms, I tried to calculate this myself and ended up with something very similar but had L(-sqrt(3)*b + a) on the numerator of the fraction instead (the alpha and beta are swapped). I was just wounding if this was a mistake or have I gone wrong somewhere? Thanks again

[@OriginalJetForMe]

Do you have recommended orientations for printing the parts? I printed them all last night on my FFF printer but now I think I should redo them on the resin printer. I’ve ordered some engineering resin that should do well for these parts.

[@Koshiro_Robot_Creator]

For FFF printers, in my experience, it is generally best to orient the parts so that the direction of the force applied to the parts is perpendicular to the print layers. Also, it’s better to ensure that the axis of screw holes is as perpendicular to the print layers as possible. For resin printers, you should place the part so that the side where you want the highest accuracy is opposite to the platform side, as the platform side tends to warp. Additionally, tilting the part at an angle seems to reduce warping.

[@OriginalJetForMe]

@@Koshiro_Robot_Creator Thanks. When I first loaded your STLs they were in all manner of orientation, so I arranged them to be more-or-less as you describe, but I'm not happy with the results for some of them. I wondered if maybe your STLs were in a specific orientation designed by you.

[@Koshiro_Robot_Creator]

​@@OriginalJetForMe Since the model was only downloaded after adjustments in Blender, the print orientation and the model's orientation in the STL file are unrelated.

[@thom7440]

Thanks for the explanation. Unfortunately it was wayyy too fast. I would have to watch it probably 100 or 200 times to break out all the details you covered to see if i actually understood what you said. I appreciate at least that you explained the details even if it's in a super condensed format. Thank you

[@yunosciences441]

thank you !

[@joelsoncdma]

use ping pong ball to slow dow ...your is hevy. Nice!

[@raphaelk00]

hello but that's a strange question, I became aware of your KZbin channel a long time ago and wanted to ask if you already have a Discord server or if you would be interested in a Discord server? if you already own a discord server I would offer to support you

[@Heisenberg2097]

If you proof this robot can also balance chicken eggs... I'll buy seven right away.

[@JensGulin]

Would you want your eggs sunny side up or over easy? 😊

[@Heisenberg2097]

@@JensGulin Please use the sAIviour of mankind to let the machine figure it by just watching my mood in the morning!!! Every good house wife can do that since thousand of years...

[@Heisenberg2097]

And if you add the original voice of the swedish chef of the muppets while frying the eggs I'll invest all my IT millions in your startup! And we'll need many millions just to get the rights from Disney!

[@araragimakoto1415]

Japanese?