Discussions

Developer

Loiter PIDs

Posted by Jason Short on October 5, 2011 at 11:18pm

I built a quick sim today to look for the optimum PIDs for Loiter:

What's looking good is

Loiter_P = .5 or .6 (more = some overshoot, but not much)

Loiter_I = zero (Just say no)

Nav_P = 4 (Higher than I thought, What will it do to navigation?)

Nav_I = .25 (Higher than I though, but still large like it should be)

Let me know if anyone proves this out!

Jason

http://wiki.ardupilot-mega.googlecode.com/git/images/copter_wp.swf

Here's why Loiter_I is a bad idea. It just ends in overshoot.

Here is Loiter_P from .5 to 1.8 (lower looks better)

This is Loiter kp 1, Loiter kI 0, Nav_kP 2 through 10, Nav_kI .2

Here it is in action:

You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –

Replies

Amit November 20, 2011 at 4:54am

on the 93 version, NAV_LOITER_I = 0 and its still overshoots, the must be a finer range parameters
Todd Hill October 13, 2011 at 6:59am

flew four waypoints last night. After hitting #4 would not RTL. Planned new mission with two waypoints, same thing. Attempted auto takeoff, copter speeded to waypoint 1, i flipped to stabalize but couldnt recover. throttle was set at minimum (my fault). flying Y6 with 750kv motors, running 2.0.47. stabilize great, alt hold so so, loiter radius too big and fast.

2011-10-13 12-34 1.log

2011-10-13 12-34 1.kmz
afernan October 12, 2011 at 10:32am

Xbee problems on 2.0.47

I've been flying with Xbee connected in order to tune Pids easy. But i found an importantnt problem: from time to time the four motors are stoped by 1 second or so, then continues normally. This happens randomly (I can't identify any reason). I've verified in my two copters, and do the same.
When I remove the Xbee, the problem dissapear. This never hapen in previous version (i don't remember which one exactly)
Anyone had this problem?
Martin Szymanski October 11, 2011 at 6:00pm

There can be integral windups.

This occurs if a disturbance such as a wind gust occurs and the steering pressure against it cannot overcome to reach a waypoint. I am not sure of the pid method used with the arducopter. I presumed that the Imax is the limit that stops integral windup?

Here is a video if two different pid methods.

The disturbance would be interpreted as the wind pressure in the copters case.

http://www.online-courses.vissim.us/Strathclyde/WindupMovie.wmv

http://www.online-courses.vissim.us/Strathclyde/WindupMovie.wmv
druckgott October 11, 2011 at 11:52am

Here are my log for .45, but loiter works ok, but older versions works better. Now I have a field 1x9 meter

With .33 it works much more better:

http://diydrones.com/forum/topics/2033-works-very-well-in-loiter

With .47 the copter totaly overshot can´t use loiter

What is the rpm limiter in the pid tuing flash file??

I do not understand the value???

2011-10-10 07-01 44.kmz

2011-10-10 07-01 44.log
Owen Booth October 11, 2011 at 2:31am

Hi Jason, I don't suppose I can get you to answer my question at this post?

Sonar attitude compensation

Thanks man, your work is awesome, and we all appreciate your efforts

regards,

Owen
Martin Szymanski October 8, 2011 at 8:56pm

There were some great flights today under Auto mode.

Integrals on both Nav and Loiter were set to .001 There was no wind.

What I observed was each time the copter began its approach to the next waypoint once it hit the maximum speed it rolled back to a stop. Then forth again in a wave like motion. Once at the derived waypoint it stopped pointed to the next waypoint with the same approach speed going up and down. Is there a derivative toward the speed m/sec, mph? The waypoint distances are small 100 m point to point. Possibly the GPS is not accurate. However i've found a garmin handheld accurate with groundspeeds. Also the IMU had to be rebooted after configuration changes were sent from MP otherwise the sonar was disabled. Attached is the log done today. MP 1.0.79 w Arducopter 2.0.46

Flying with no wind is enjoyable.

2011-10-08 1.log

https://storage.ning.com/topology/rest/1.0/file/get/3692282751?profile=original
afernan October 8, 2011 at 2:18pm

Jason,
I was testing a lot tuning the loiter pids with fw.46, trying to fix my "circles" problem. I've fixed finally with pids close to the ones you propose now by default. So, i reduced Loiter_p to 0.6 and circle effect almost dissapear. Then i increased nav_p and nav_i up to 2.7 and 0.27 respectively. I've got this good performances only by testing a lot, before reading this post. Now the behaviour is much better. Tomorrow I'll test fw.47
Amit October 8, 2011 at 10:18am

On the 47 ver i have set the compass to the north, but when i rotate it 180 to the south there is 20 deg differences for the real south.

Any clue?
Balloon October 8, 2011 at 2:32am

Jason,

It have problem last code 2.0.47 . when i will change mode (ALT or Loiter) and I change back to Stabilize mode. I can't Throttle control and slow down on ground (No crash). and I down throttle and up again. I just normal control. Now it's problem about throttle when I will change mode to ALT or Loiter ..... Edit Now!! Please........

thank you , Balloon