When selecting an actuator for a controls application - e.g. choosing a servo motor for a self-balancing robot - I understand that it's important for the actuators to have fast enough dynamics that they don't significantly impact the system's linearity, and thus ensure the control algorithms function as designed. However, how much faster is fast enough? Are there any rules of thumb - c.f. Nyquist's Theorem in signal processing - for how I should choose my motor's rise time given that my controller's sample period is known?

  • $\begingroup$ The actuator response time should be compared with the response of the system, not on the controller sample rate. If the uncontrolled robot falls over in 1 second, it makes no difference to the actuator that is required if your control system uses 10 samples / second or 10,000. Using 10,000 samples per second doesn't mean you have to "eliminate every position error" in 1/10,000 of a second with a vastly over-designed actuator! Unless you are assuming your actuators are "perfect" devices, their own dynamic response should be included in the system model anyway. $\endgroup$
    – alephzero
    Jul 11 '19 at 11:09
  • $\begingroup$ Ah, that's a much more sensible way of framing the problem. And I was wondering why I couldn't find anything in the textbooks... Thanks! $\endgroup$
    – jd9610
    Jul 12 '19 at 2:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.