Hey, so my team is looking for ways to make the RPM function to be more precise. We are currently using linear interpolation in Excel. However we feel that probably a more real function could be achieved, and thought of calculating a parabola. How could we possibly achieve that, and is it worth it?
You want a polynomial regression of order 2.
P.S. if you weren’t finding answers on Google, it’s spelled quadratic, not cuadratic
To what RPM function are you referring?
In any case, to do a smoother interpolation function that actually passes through your data points, you may want to try cubic splines.
I’d recommend using desmos Graphing calculator. It has a built in regression calculator with any equation type you want. We use it for regressions all the time and it’s super easy to use.
2 Likes
I second this. Desmos has a very powerful regression tool. You can run a regression based on any type of equation, and can run multivariable regressions.
I actually found/implemented code to calculate cubic spline interpolations on the reo. The code is here https://github.com/phlogiston1/frc-3461-2020/blob/master/src/main/java/frc/lib/math/CubicSplineInterpolate.java you do need the Jama libraries for matrices but if you take them from the folder on our GitHub it only has the parts required for the interpolation.
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.