How to find kS and kV values without sysID (2024)

[duje]

How do we tune kS and kV values without using sysID? There doesn’t seem to be support for finding them in the 2024 application

[vargoose84]

I think you might be mistaken.

Did you visit the docs:

Introduction to System Identification

Or search CD:

Ok don't panic! SysID is "half" going away!

Our team is super appreciative of WPILIB and all the work that they do. Without the cool infrastructures and tools like command-based programming, and vast libraries that actually make “talking” to the components of our robot feasible, I’m sure our team would not have been as successful as we were. I say this so that no shade is thrown at the wonderful WPILIB team. Soooooo, when we heard that SysID is moving half into the robot code and half an offline tool we started to panic. You see, we a…

[ArchdukeTim]

I don’t think he means there’s no support for SysID. He means there’s no support for calculating the system characteristics without SysID.

OP, you could try getting ‘close enough’ variables by plugging your mechanism details into ReCalc. YMMV though

[rafi]

Fundamentally, all you need to do to find kS and kV is run a test where your acceleration is negligible while capturing as wide a range of velocities as you can. The easiest way to do this is to run a voltage ramp, where you start by outputting 0V and you slowly increase the voltage output. This way your acceleration is very small, but you can plot your velocity vs voltage. If you put voltage on the Y axis and velocity on the X axis, you should be able to do a line of best fit where the Y-intercept is kS and the slope is kV, matching the feed forward equation V = kS + v*kV.

SysID just provides you with tools to collect data and perform a more robust version of this analysis that can also compute kA (if you get the right data).

[vargoose84]

@duje

There doesn’t seem to be support for finding them in the 2024 application

I mistook this statement to mean that the op wanted to figure this out empirically, my bad if that’s not the case.

[asid61]

Put simpler, you can just take a few points of voltage vs final settled speed and plot the result. The X intercept is kS and the slope is kV.

[viggy96]

You can find the other feed forward values using recalc, and kS can be found empirically by finding the smallest voltage that makes the mechanism move.

How to obtain kS for feed-forward

What’s the best way to find kS? Can it be relatively easily found like tuning for kP/kI/kD? I believe that I should be able to use Recalc to get the other feed forward values (kV, kA etc).

[gerthworm]

experimental method:

1. set kS to the largest value you can without seeing any motion in the mechanism.
2. start with kV = 12.0 / mechanism max speed . Iterate up and down until the mechanism goes the commanded speed.

[calcmogul]

Alternatively, for step 2 use kᵥ = (u − kₛ)/x, where u is some voltage and x is the steady-state velocity measured at that voltage. That way, you only need to sample one point. Granted, it won’t be as representative as using multiple samples and a line fit.