Servo 'smoothing'

[EricVanWyk]

Originally Posted by Kevin Sevcik

Erik,

Translating your IIR to fixed point doesn't seem entirely straightforward to me. Or, at least, we should remind the audience of the wonderful effects known as truncation and loss of precision. You'd need to process your new value and carry your old value as scaled integers, or you'd lose everything behind the decimal point and get weird jumps in your values. So it'd look something like:

Kevin - You are 100% correct in calling me on missing the target audience. Sorry all!

You will have to forgive me, I was working under the assumption that the processor had a built-in 32bit FPU. I'm so glad that >> and << optimizations will soon be a thing of the past. I started on the old PBASIC systems, and quickly learned that some optimizations simply obfuscate code. Now I make sure that the first pass is as easy to read as possible - sweet beautiful pseudo code. The second pass can compile.


Your translation is correct, I think. For those following along at home, Kevin moved tau from the 0..1 domain to the 0..15 domain. This allows him to use fixed point math and the shift operator. I'd move it into the 0..255 domain so you can use a wheel input as tau. If you do this, use 16 bit numbers.

[nickmagus]

I've had some experience with jittery pots myself i find the first thing you should do is debug every single thing in that system until you find out exactly what is failing. If your assumption is correct then i suggest you map the jitters and see if your digital position vs actual position follows a pattern. if it does than there is the easiest and best way to fix the problem remap the digital value with a lookup table to match the actual value. if you find that the jitters are indeed white noise and are completely random then i suggest a filter such as the ones proposed above (a simple average of the last few values should work fine.) keep in mind that jitters can be caused by all sorts of things not just bad pots i had a magnetic encoder i thought was bad and it turned out that the bolts just above it were causing it to "jitter".