PAROL6: 3D-printed desktop robotic arm
source-robotics.github.io
source-robotics.github.io
Check out the SO100 arm, being supported by Huggingface and others. Only 4 axis, but cheap AF (<$250/arm) and using ML to make it more capable than 4 axis would seem. Also using identical arms for mirror teleportation/training.
Perhaps 4 is enough for any specific application, but then again perhaps 3 is or 2. :P
I've never programmed a robot arm, but I've spent a fair amount of time using a seven axis faro arm (a coordinate measuring device, sort of the opposite of a robot arm) and it certainly takes some practice to avoid "cant move there from here without reorienting everything", it's easy to take for granted what our brains do automatically for us. :)
There are a lot of downsides in 'arms' -- stacking all the axis hurts their speed, stiffnes, accuracy, increase complexity (e.g. in pathing, limiting speed, restricting to safe areas), limited reach.
I've always thought of them primarily as beneficial for flexibility at considerable cost, but if it's not 6dof then the flexibility isn't so great, then why not some other geometry?
Drone and helicopters has 4, and they are able to control max 4 of 6 parameters. Usually 3 positional + 1 rotational, and the rotatinal axis go first.
BTW, looks like it doesn't have closed loop control. "0.2mm repetability" (should be repeAtability) is only 'under certain conditions', no load.
https://www.youtube.com/watch?v=ECLrLupFW10
https://www.youtube.com/watch?v=_4mrb2T706s
https://www.youtube.com/watch?v=Ctb4s6fqnqo
https://github.com/adamb314/ServoProjectThe PAROL6 BOM doesn't provide any cost estimate, but it looks quite a bit more expensive.
Also they do measure position to achieve their feedback, might as well just output that on a 4th wire.
So when you use hobby servos for robots, you're taking a low precision actuator meant to make a flap or throttle or other control surface go relatively up, down, in, or out, that was designed in like, the 70s, and asking it to do modern robotics stuff
If you think about it, it actually makes a tiny bit of sense. First, it is failsafe: Breaking the control line or shorting it to ground will not move the servo to 0%, shorting it to signal level will not move it to 100% - it just doesn't move at all and stops applying force. Any sentient being within the movement range will definitely prefer it that way instead of random movements. Second, it can actually be pretty precise: The driver circuit can be completely analog, it doesn't have to be limited by arbitrary digital quantization steps. All it needs to do is check if the current encoder value is above or below the target and apply power to the motor accordingly.
Some local people made this with it: https://youtube.com/watch?v=J3BygZpa6Eo
https://github.com/AlexanderKoch-Koch/low_cost_robot/tree/ma...
Ok, now I REALLY want to build one...
1000e kit does not include steppers and board..
In https://source-robotics.github.io/PAROL-docs/page6/ it tells you it can pinch you.
It's probably relatively safe for humans though (be careful with your pets), the robot is 5kg, it move quite slowly and motors are nema 17 (the same than in a 3d printer) with a max gearbox ratio x20. In case of problem press the e-stop button if you have chosen to have one.
As far as I understand there are no encoder on the joints so it can't be back-driven, it's not a compliant robot.
Rant mode activated : Those prices are just excessive. Before 2020 we could buy some 5-axis 8-bits RAMPS 1.4 board with stepper for less than 50€. Then there was some transition to use some more powerful 32-bits microcontroller (which are cheaper on the BOM), but no standard has emerged, 32-bits chips were impacted by the shortage, so everyone and their dog are creating their own 3d printer control boards, and selling them for more than €100, and they are less customizable, harder to upload. (The best replacement for RAMPS 1.4, is probably BTT SKR V1.4, which sells for €43 with silent drivers) (but you are on your own for the non-standard software and development environment).
Those control boards are just connector boards, they are mainly used to replace some wires, it's just to connect the microcontroller to the stepper driver. You can have your own pcb connector board made for less than €10 for 10. You just have to solder the header yourself.
The 32-bits transition is better and cheaper on the BOM compared to the Mega2560. The speed boost is really useful when you want to add more advance control algorithm for your motors for robotics application.
Creality CR-10 quality was just better and it was the beginning of the transition away from open standards. A slow but sure erosion of what was once a great flexible community solution, where what was built by open-source is being privatized.
Chinese clones replicated CR-10 model each with their own control board. But when your machine is repairable, you can't sell it for a lot more than the sum of individual parts, so the whole economics of how much a control board that run marlin is got delirious.
This trend continue today with the emergence of companies like BambuLabs.
The whole arc makes me sad, as 3d-printing is the entry point for many makers around the world and these initial core values of really owning your hardware stack are important. For building robots, building your own 3d printer was a great introduction. Being able to replace a fried motor driver because your robot took an unexpected fall for cheap is kind of essential. If every time a part break you need to pay more than €100 and more than to 2 weeks of delivery time, you can't test and push your machines. Whereas if you are using standard parts, you'll have a stock of them, and it will be €5 and 5 minutes and you are good to go, and the mistake is forgotten, and the lesson learned.