Differential_drive PID controller bug?

asked 2015-07-19 08:11:28 -0600

miguel gravatar image

updated 2015-07-19 08:13:59 -0600

Hi everyone,

I am having issues tuning the PID parameters for the differential drive package. I see on the external site there is a mention of a few bugs in the PID controller. The controller tends to work correctly but sometimes reacts very slowly after a few seconds. My motors often stall and the controller takes a long time to increase and overcome the friction of the wheels. Ive tried to increase the integral gain but it still takes too long.

I am not a controller wiz so cant really debug the program. Has anybody used the package and find any bugs?

Also, does anybody know what the rolling_points parameter means? should it be tuned or adjusted?

Any help will be greatly appreciated.

edit retag flag offensive close merge delete

Comments

The knex_ros site is not clear about whether the bugs mentioned in the PID controller are a tuning issue or implementation bugs in the differential_drive package.

ahendrix gravatar imageahendrix ( 2015-07-19 12:01:10 -0600 )edit

Do you have any suggestions about what I should do?

miguel gravatar imagemiguel ( 2015-07-20 05:25:20 -0600 )edit

I'm not familiar with the differential_drive package, but I think you should spend more time trying to tune your controller and less time looking for bugs. If you're seeing lag creep into the system over time, you may want to try decreasing the controller loop rate.

ahendrix gravatar imageahendrix ( 2015-07-21 01:42:43 -0600 )edit

Initially I did try tuning the controller but was having some issues with that, hence the reason for the bug investigation. Is there any concrete way in tuning the gains? ill try spend some more time again.

miguel gravatar imagemiguel ( 2015-07-21 02:07:37 -0600 )edit