1
$\begingroup$

I'm building a 7oF arm with series elastic compliant joints. I need series elasticity because I'm training the robot to do autonomous manipulation using machine learning, and I want force control (and prevent the arm from killing itself if it hits something). I want the final joint to be something like this.

I'm having real trouble finding plans online for any kind of rotary series elastic joint. I wonder why this is, given how easy it is to find arm plans.

I'm going to add it to this robot. An example of the kind of motor I'll be sticking it on is 5:1 planetary gear stepper motor.

My plan right now is to buy a rotary magnetic absolute encoder and a torsional spring. I don't really know if there's a standard way to mount these things.

So if anyone could point me in the right direction, that would be great :)

$\endgroup$
2
  • 1
    $\begingroup$ Unless you are a researcher hopping on the buzzword bandwagon, you should have a specific reason for adding series elasticity. If you add a bit of your reason to your question above it will help get better answers. $\endgroup$
    – hauptmech
    Commented Sep 28, 2017 at 2:38
  • $\begingroup$ @hauptmech thanks -- is this OK? And if you could point me to some info on the correct procedure for designing robotic arms (designing structural framework, what materials you'd typically use, designing hardware controllers), then that would be much appreciated! $\endgroup$
    – Tarrare
    Commented Sep 28, 2017 at 8:03

2 Answers 2

1
$\begingroup$

For an arm with series elastic actuators I'm not aware of an open source design. There are many mechanisms you can find in research papers if you search for them.

For force control you have a few other options.

You can use an efficient backdrivable transmission (direct drive or spur or planetary or capstan all work) and control the current in a permanent magnet motor which will have a linear relationship between current and torque/force.

If you only interact with the world (payload) at the end-effector, you can put a force-torque sensor there.

For safety you can use clutches instead of compliance in many cases.

In all cases you will need to invest some time and learning. Maybe buddy up with a mechanical engineer.

As a side note, there are cheaper ways to measure the deflection (the original used a potentiometer) and finding the correct geometry and size of spring to match the operating conditions is a challenge in itself. These guys used compression spring for a rotary actuator and you could probably find enough public information plus some simple equations to use that approach in your own design.

A final comment is that since you tagged visual servoing in your other question and your are talking about machine learning, you might consider keeping things more simple. Don't worry about force control, add whatever compliance you need for safety in any form (a rubber section in the link perhaps) and let the learning algorithm handle the control details...

$\endgroup$
3
  • $\begingroup$ Great answer and I really appreciate you writing this. What books/courses/projects would you recommend to get a good grasp of the basics? $\endgroup$
    – Tarrare
    Commented Sep 28, 2017 at 11:19
  • $\begingroup$ No recommendations other than if you're serious about robotics, learn from researchers, industry specialists and academics rather than hobbyists. You'll go further faster that way. $\endgroup$
    – hauptmech
    Commented Sep 30, 2017 at 1:22
  • $\begingroup$ thanks. Is there a forum (apart from here) that they hang out in? $\endgroup$
    – Tarrare
    Commented Sep 30, 2017 at 7:39
1
$\begingroup$

There was a paper describing a rotary series elastic actuator for human-assistive systems in IEEE/ASME Transactions on Mechatronics, Vol. 17, No. 2 (April 2012).

The paper included diagrams of the mechanism and a description of the controller design.

Is that the sort of thing you were looking for?

$\endgroup$
3
  • $\begingroup$ Thank you for the link, this is really good. However, I'm really looking for something at a more pragmatic level: a parts list and STL model, eventually resulting in something that I could just drop into place of an existing servo with minimum modification to the external structure. Is this the right place to ask? $\endgroup$
    – Tarrare
    Commented Sep 27, 2017 at 14:13
  • $\begingroup$ e.g. this is better, but it's not inline so I couldn't drop it in place of an inline servo. $\endgroup$
    – Tarrare
    Commented Sep 27, 2017 at 15:18
  • $\begingroup$ Or something like this... but it has the same problem, can't drop it in place of an existing servo, and would have to machine the torsional spring myself $\endgroup$
    – Tarrare
    Commented Sep 27, 2017 at 15:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.