Allow different PID values for pre/post rotate and target vs max error #3
Conversation
While FOLLOWING angle PID needs to be weaker so that lat PID dominates, in PRE/POST rotate it needs to be stronger. Also make the max error the maximum, not the target, to allow the integral error to build.
| grp_pid.add("kp_ang_rotate", double_t, 0, "KP for angle in pre/post rotate", 1.0, 0.0, 50.0) | ||
| grp_pid.add("ki_ang_rotate", double_t, 0, "KI for angle in pre/post rotate", 0.0, 0.0, 5.0) | ||
| grp_pid.add("ki_ang_max_rotate", double_t, 0, "KI windup for angle in pre/post rotate", 10.0, 0.0, 500.0) | ||
| grp_pid.add("kd_ang_rotate", double_t, 0, "KD for angle in pre/post rotate", 0.0, 0.0, 5.0) |
There was a problem hiding this comment.
Is it intended that these values are the same as those without _rotate? So different settings would be possible, but not used by default?
There was a problem hiding this comment.
Exactly, didn't want to change current behaviour but I'm using higher values
There was a problem hiding this comment.
Which values do you use then? I'll see if the target vs. error helps, but if not, it would be good to know which values to increase how much.
There was a problem hiding this comment.
I'm using completely different values as I've changed how the motors are controlled.
Looking at https://github.com/ClemensElflein/open_mower_ros/blob/main/src/open_mower/params/ftc_local_planner.yaml
I would start with kp_ang_rotate: 3.5, ki_ang_rotate: 0.5
2 participants
While FOLLOWING angle PID needs to be weaker so that lat PID dominates, in PRE/POST rotate it needs to be stronger.
Also make the max error the maximum, not the target, to allow the integral error to build.