Robot with DOF +-Z, +X, ±𝜃y

  • BI1WhFIHello,


    I am brainstorming for a robotic solution to a particular type of motion for an optical application and would love to hear any suggestions or links to products that may aid in my efforts. Please see my image for context in the upper right corner.


    I have an optical assembly that rotates on a disc in a 50 cm radius in the XY plane. It needs to have out of plane motion such that the stage follows a 50 cm radius arc in the XZ plane, and the optical assembly still points at the origin (hence the "tilt" motion (DOF in theta_y)).


    I have some obvious ideas such as a simple linkage mechanism and a linear actuator; however the purpose of this brainstorming is to see if there exists some suitable OTS robotics product that can maneuver an optical stage in the required arcing motion independent of some linkage system.


    The required DOF would be +-10cm in Z, +X (to position the stage along the arc), and a tilt about Y (theta_y). In the tilt actuators I have seen, they don't have the capability to translate such a large height, and in both directions. Hard to put a number on the weight of the optics, but for the implied footprint in the image, I imagine its on the higher capacity spectrum for suitable robotics. I'm guessing somewhere between 30-50 kg.


    I am looking for elegant solutions. Given that some aforementioned linkage system could perform this motion with a single DOF (+-Z), the robotic stage approach would require a fully contained actuator to "compete", and not some composite configuration with many stages which would complicate the system.


    Thank you,

    EB

  • Place your Ad here!
  • Hi HawkME,


    I would say definitely sub-mm, probably on the order of 100 um.

    Is that repeatability, or accuracy? Because 100 microns repeatability is achievable for "tighter" robots, but isn't going to be achievable in accuracy without taking additional steps.


    For example, this is part of the spec sheet from a KUKA KR60HA, which is a "High Accuracy" robot:


    Also, note that the RP value is radial -- some people quote their accuracy/repeatability as a diameter (often "True Position"), which can lead to unfortunate misunderstandings.


    And never forget that "spec sheet" values are attained in laboratory conditions -- real-world production situations are less perfect.


    To be clear, I have achieved 250 micron accuracy with KR60HAs in production scenarios, but it took a substantial amount of tuning. Basically, after mounting the working payload, we did a real-world 6-DOF mapping of accuracy errors throughout the working volume, and built a correction table. The good news is, there are off-the-shelf products, like New River Kinematics' SARCA module, that can do this for you.

  • Hi SkyeFire,


    Thanks for the reply. Unfortunately I simply don't have the detailed requirements for accuracy/repeatability because that is dependent on the optics of the system which are not fully defined yet. I am more concerned with simply finding an actuator/robot/stage that even meets my specified DOF, geometric, and loading requirements before I become too concerned with the accuracy/repeatability.


    For example, one interesting actuator I see online is a motorized goniometer which could theoretically provide the required arcing motion, but I have yet to find one with a large enough arc radius and load capacity that would make sense. Also the optical stage would need to be mounted perpendicular to such an actuator which could get messy.


    Another thought is some sort of XY-theta gantry turned on its side so the XY motion is in the XZ plane. But this just sounds complicated already which is something I want to avoid.

  • Well, a 6-axis arm would almost certainly give you the range of motion required -- it would simply be a matter of picking the right model. But the Accuracy/Repeatbility question could be a deal-breaker.


    The "deluxe" answer would be a custom one-off system, built to your requirements, but that's going to be $$$ and create a longevity issue -- one-off machines tend to be a bit hard to find tech support for after a few years.


    There's also the question of whether this will be a "static" or "dynamic" application -- that is, do you move to each position, come to a complete stop (plus settling/ringing time), then perform your optical operation before moving on? Or is this supposed to be some kind of "flying measurement" thing? B/c the later is going to be much more difficult and/or expensive to get made and working accurately.

  • Hi SkyeFire,


    Good points. This application you would call a static application where an optical measurement is taken for some length of time while the robot is positioned at some coordinate, and the moves directly to the next position to take the next measurement.


    After some internal discussions, we are leaning towards a custom solution that adds a minimum number of degrees of freedom.


    One thought is some kind of 3 axis Stewart platform, although I'm not sure if its possible to construct one in the traditional sense and achieve the requested DOF. It might have to be some serially stacked robot. An advantage of the Stewart platforms are the very high repeatability due to the parallel actuators acting on a single, small load.


    Yes a 6 DOF Stewart platform would probably work, but it is expensive, overkill, and not totally tailored to our system. I'm still interested in some type of robot with 3 DOF to achieve exactly the DOF I need.

  • I wonder if a SCARA robot might serve? SCARAs provide XYZ and RZ, so to get RY, you might need to mount it sideways on a wall (always check to see if the robot can support that, some only work with a particular relationship to gravity), but SCARAs tend to be higher accuracy due to their lower axis count and "tighter" build.


    A SCARA with an arm long enough could probably generate your 50cm-radius motion arc, and use the Z stroke as one of your required translation axes.

    External Content youtu.be
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account
Sign up for a new account in our community. It's easy!
Register a new account
Sign in
Already have an account? Sign in here.
Sign in Now

Advertising from our partners