วีดีโอ

@@АнтонСергиенко-ь2ц its a very interesting mechanism, but there is very little information about it!

@@swannschilling474 Well, there are a lot of videos on this topic on TH-cam, some of which show the structure of such a mechanism in quite some detail.

@@АнтонСергиенко-ь2ц I also saw a few, but not really anything on how to model it, or on the precision or torque of this joint! But it seems to be having a super high workspace, which makes it very interesting!

@@swannschilling474 It's strange, I come across videos with detailed explanations, except that they don't provide drawings) Of course it's very interesting, but there are two downsides - it can't be used everywhere due to its dimensions and problems with force application.

@@swannschilling474There is also such an interesting mechanism, but it seems that it is also not for power load. th-cam.com/video/BkjTnm2bnH0/w-d-xo.htmlsi=FcO3pId2GVy28oGy

github.com/SwannSchilling/SPM_Kinematics This is the github page, let me know if you need more info...any help solving this problem would be great! 😊

🦾

@@dirkdiggler2430 Buy but since they are BLDCs you are not missing much ASMR!! 😁

Yes it is slightly different because it uses standard encoders, the ones in the video are custom encoders! The rest will be the same as in the video! Those motors are out of stock, I will have to make another version soon, which is using available motors! Sorry about that! Sometimes AliExpress is not the very best option!

Which bearings are you talking about? Let me know if I need to fix a mistake in the BOM!!

@@swannschilling474 3 bearings on the 3 shafts of the 3 motors. then we see the first SPUR mounted using no extra bearings. then the other two spurs get mounted with 1 bearing each. so in total of 5 bearings are being used here instead of 6 as mentioned?.... am I right?

@@SaintAngerFTW its three bearings on the spur gears first layer, then two on the second layer and one on the third layer...it might be a little different because the uploaded printing files are having the encoders sandwiched between layers, rather than on top!

I am glad that you like it! I hope I will have the time to build V2 sometime soon! 😊

That would be a very good usecase! Never thought about that! 😊

@@swannschilling474 the idea might be superseded by a mass producible 360° Infrared space telescope concept.

@@Astroponicist wow, you seem to know a lot more about this topic than I do! I was mainly thinking about using this as a joint for my robotic arm...

@@swannschilling474 your device might be a good steady cam mount. I am available for consultation anytime.

@@Astroponicist steady cam mount was actually also in my mind...there are still some caveats, one of the biggest issues is that I am running a simulation to drive it, and have not figured out the kinematics for being able to drive it in a more generic way! Wanna give me a hand figuring it out? 😇

Thanks!! It was really fun to build!!

@@swannschilling474 is there someone I could message you ? I had a few questions about the motors your recommended

pip install jupyterlab-urdf And it will work like a charm! 😉

A gimbal would be a very good usecase! 😊

That is actually a legit point, for something with a wide base this would cause a problem! But I build my gripper in a way that there is enough room, so in my case there are no issues! There are a lot of SPM design possibilities out there and it is always a good idea to adjust them to your needs... and yes one of mine was to make it look cool! 😁

@@swannschilling474 Well it does look cool for sure, so good job.

No, its Open Source!!! All yours if you'd likebtp build one! 😊

Hahaha did you manage to put it together?

Let me know if you did!! 😁

Hahaha, glad you like it!! Its my favorite screw driver!! 😁

Very happy that you liked it! Unity was about to create a lot of content for the use in robotics, sadly it got discontinued! I am planning to move the arm with a virtual overlay using AR on the Quest 3 by deploying my Unity simulation...its in the making, stay tuned for more! 🤖

I am using it as end effector for my robot arm..you could use it as a gimbal, or a centered rotational point when observing things with a microscope etc.

@@swannschilling474 Someone else did a video on using it for microscopy. That's what got me to yours. I was going to comment on your other video that I was really hoping you'd use the accelerometer/gyro in the ps4 controller to mimic the orientation of it but I have no idea how nicely Unity on linux plays with PS controllers. Usually everyone says "get an xbox controller instead". Do you think Unity is the best choice for linux? I have a lot of things I need visualization for but so far I've been using the xorg libraries straight. Its a PITA. I've been considering Godot.

Thanks so much, I was too busy to push this project forward...but I might actually put more work in soon! 😊

I actually never figured out how to setup the URDF of this joint, because of the closed loop kinematic, I would love to try it in Pybullet! But honestly Unity and either just sending it directly to COM or to a Flask server is just such a nice and easy thing...it rocks!! 🤘😁

Thank you so much, I wish I had the time to work a bit more on this!! 😊

This is because the body used to generate the circular pattern is included in the tools collection and it's later used as the target body in the combine method. A simple hack is to add a line to remove the target body from the tool bodies collection (last two lines below). ... for b in circularFeat.bodies: ToolBodies.add(b) # add this line to remove target body ToolBodies.removeByItem(body1) ... Edit: remove(body1) changed to removeByItem(body1)

Hey, sorry for the late reply...it seems like you are passing invalid arguments to the script! Please try to use the same settings that I used and go from there! If you are running into additional issues please let ne know! 😇

@@swannschilling474 I believe it's because the target body of the combine action is included in the tools collection. I had to add a line of code to remove the target body (ToolBodies.removeByItem(body1)) to get it to work. I'm sure there is a more elegant solution.

@@swannschilling474 hi i use how is your script without any changes the cylinder must by how you have it also ?

@@TasosManolidis It has been some time since I used the script...I will check if it still works or if it is a Fusion 360 being updated problem!

Thank you so much! The files for printing are available if you're interested to build one yourself! 😇

Thank you so much!! 😊

This is still in developement but I will try to show as much as I can about how to set it up in the future!

The problem is that the motors got extremely expensive...I will try to make an updated version using different motors in the future! 😊

I am using an ODrive, since the motors are BLDCs...I am planning to do more content on the robot arm soon! 😊

It is just the ratio of the gears attached to the motor axis and it corresponds 1:10 in ratio to the disks number of teeth! 😊

They are just little inserts that hold the bearings im place...the design works well, but I am about to design V2. Let me know how everything is working out for you! 😊

I made it using ai tools but its not completely ai generated, it is actually the first time I managed to get something that I thought it might be fun to share! I hope you liked it! 😊

Automated TH-cam comments to drive engagement?

Funny! I was looking at your channel this morning! 😇 There are a few things that I would like to change...but the gears are working fine actually! I am trying to figure out the inverse kinematics atm and I am stuck! Did you get the math done?

@@swannschilling474 it's been a bit since I've worked on it. There is a good paper I may have... somewhere but the short answer is no, I don't have the kinematics worked out yet.

@@colbertbd let me know if you find that paper! I am having a github repo, if you would like to take a look you can check my progress. I am kinda stuck on the inverse kinematics... github.com/SwannSchilling/SPM_Kinematics

OMG!!! What has been heard cannot be unheard...happened to me just now!! 😅

Sorry for that, I hope you still enjoyed the video! :)

You should assemble everything in Blender first, Unity is not made for this!! Just find a tutorial on how to set pivot points in Blender and you are all set...its not a big deal! 😊

@@swannschilling474 But when I import the STL files into blender, the model parts are not visible. They are all apparently at the origin, so they are not in the correct positions. Any ideas?

@@richc2419 Select the object then View>Frame Selected But I would advice you to learn the basics of Blender!

@@swannschilling474 Thanks, I did actually figure that out -- the models are off-center, but there are a couple meshes separated from the others, not sure where they go, exactly.

@@richc2419 The assembly video has all the parts included, but I could also upload the mechanism as a single object for reference? 😊

They are released, I will put the link to the description later! 😊

3D Printing Files www.printables.com/model/323998-spm-3d-printed-spherical-parallel-manipulator

@@swannschilling474 Awesome!

@@ghostdog0424 Soryy, that I forgot in the first place!! Glad you like it!! ☺

@@swannschilling474 Can't wait to try this out!

It is ABS, I was planning to used Carbon Filament but got sidetracked by LLMs...will get back to robotics soon though!! 😅

Thanks!!

Yes everything of the inverese kinematics is calculated inside of Unity, but I have not touched this project in a while...might eventually go back to it soon though!

​@@swannschilling474 Thank you so much, I really take care of this project and I looking for more information about inverse kinematics. Your project is the inspiration for me. Hope you come back soon.

@@NgọcTuấnNguyễn-w8i glad to hear you are enjoying the project! Its all Unity based for now, so maybe getting into Unity might be a good starting point if you'd like to build something similar!

The 3d files are published! www.printables.com/model/323998-spm-3d-printed-spherical-parallel-manipulator

@@swannschilling474 thank you

Thanks a lot, I want to keep building for this project again soon!

Thank you so much!! 🤩

I had to ways to do this, the first one is using articulation bodies. So the whole mechanism is basically being simulated...they other one is an IK rig. I used the IK rig in the end, because it was more stable than the simulation, but both ways are equally legit!

@@swannschilling474 thank you very much for your answer. This helps me very much

@@duongTran-he6mk No problem!! Have fun building...let me know if you would like the link to the .stl files!

@@swannschilling474 Thank you very much for your kind suggestion. I've already had my STL files. My problem is with the Unity simulation. I used the Articulation body but don't know how to deal with parallel model like SPM, because Articulation body seems cannot deal with loop kinematic. I'll try the IK rig as you suggested.

@@duongTran-he6mk IK did the trick for me in the end! 😊

As far as I can tell it is just coated metal, I don't think it is ferromagnetic because the magnetic field would disturb the electromagnetic field that is being build by the copper wires...but that is just my idea, correct me if I am wrong! ☺

Its my design! 😊

@@swannschilling474 it's amazing 👏👍

@@velkam848 thank you so much! I hope I can build v2 soon! 😊