The PID Controller

I built again the model with two propellers and used the WindMill Robo Pro project file shared by winijenh in his post.

It works like a charm. 

I didn't go into the Kalman filter specifics. However, I looked deeper into the PID controller. I used the manual tuning method (reference):
  1. Set all value to 0
  2. Increase Kp till the balance starts to move and oscillate
  3. Keep half of Kp
  4. Increase Ki to correct the process
  5. Increase Kd to improve speed
It took me 10 minutes to set the values of Kp, Ki and Kd. 

Too high values for Kp, Ki and Kd cause instability. Don't hesitate to decrease them.

The different sets of values I obtained for a good stability are : 
  • Kp=14, Ki=0, Kd=8
  • Kp=11, Ki=6, Kd=11
Playing with different sets of values alloy to fully understand proportional term or gain, integral term or reset, and derivative term or rate.

Because Ki hasn't a big impact, I suggest this modified approach:
  1. Set all value to 0
  2. Increase Kp till the balance starts to move and oscillate
  3. Keep half of Kp
  4. Increase Kd to obtain correction when the device is unbalanced
  5. Optionally, increase Ki to improve the horizontality of the device

Download

The attached file is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. 
equilibre_4_works_fine.rpp.zip
File Size: 44 kb
File Type: zip
Download File

Previous

Next