|
Re: paper: Modified Bilbo bang-bang .vi
We actually used the built in labview PID control. We only used the PI terms, so it was a PI control. We had 3 closed loop shot selections, 1500 RPM for the fender, 2500 for front of key and 2800 for rear of key. While tuning we couldn't get one set of gains to work at all 3 set point rpms, so we had to employ a "gain shifter" based on the RPM requested. We also limited the output range between motor values that were expected. This also gave it some "Feed forward" as the PID was limited to a range that was expected. We tried a couple of other methods, but this method gave us the best response. We did not try the bang-bang method, but looking at trying it, as it would eliminate some different pid code between autonomous and teleop.
We are planning on doing some off season events, so looking to try this method and see how it performs.
I always thought the pid was overkill, but at the 1500 rpm we saw deviations in the +_50 rpm range. From what I read, it sounded like the bang-bang gave tighter control than what we saw with pid.
__________________
The sooner we get behind schedule, the more time we have to catch up.
|