Version 2.5 of the ArduCopter code is now available in the AP Mission Planner and in the downloads area!
Enhancements:
- The first ArduCopter code release optimized for the APM 2. Leans and drift should be much reduced or even eliminated for most users. This was accomplished through a number of core improvements to the DCM implementation by Tridge like this one and this one.
- Loiter and waypoint following should be improved due to a D term bug fix, some tuning and the improved DCM performance mentioned above.
- On start-up, the yaw heading is updated with first successful mag read (so you should no longer see the slow rotation from north to your actual heading).
- Increased output rate to ESCs to 490hz. This update rate is also user selectable using the new RC_SPEED parameter.
- hexa copter stability patch bug fix (should resolve slight flattening when pitching forward and accelerating very rapidly).
- improved baro filtering
- fix to dataflash logging of Mag heading
- addition of H1 swash plate type and bug fix for proactive yaw compensation for collective pitch changes for TradHelis.
Tuning:
The default PIDs are optimized for a 3DR/Jdrones quad with 850 motors and 10" props. If you're using more powerful motors/props, start by turning down Rate Roll P in 25% steps.
While some testers have reported very good flights with the default PIDs, some have reported that this release is a little "sharper" due to the DCM improvements and have found they needed to:
reduce stabilize P by 10% (i.e. 4.5 -> 4.2)
reduce stabilize D by 30% (i.e. 0.15 -> 0.10)
increase rate D from 0 to 0.001
Tuning loiter can be tricky. Refer to the discussions which will appear below for more community feedback on what parameters work best.
Please post your feedback in this discussion. For enhancement requests and bug report, please add them to the arducopter issues list. When possible please include logs (tlog and/or dataflash) and tell us whether you're using APM1 or APM2 and what version of the software you're using (presumably 2.5 but tell us anyway!).
Thanks for this release go to the developers (both in the core team but also those who have provided bug fixes through the issues list) and also the community members who participated in the previous release thread and provided some great detailed information in the form of issue reports and logs which allowed us to nail some bugs!
Replies
Can someone post their Loiter Numbers along with Nav numbers...
Along with any info on how to tune Loiter..
WIth Loiter_P at 3.0 my tri drifted off by 400 feet measured... but it did stayed at 10 feet +- 2 feet !!!
Thanks
Eddie Weeks
I have flown almost every version since NG and I knew in a millisecond 2.5x was a major leap ahead. I have a x525 quad, apm2
http://diydrones.com/forum/topics/share-your-drone-info-thread?id=7...
and the instant I took off with 2.53 I knew it was great. All of us who fly copters know ground effect and the need to pop up a couple of feet to get stable. This software can float the quad an inch high with no ground effect! Absolutely the smoothest thing ever.
Dave C is right - tuning is now like it should be, no longer change .001 and it is unflyable. I have used many of Marco's pids - the rate / stab tweaks really make it nice and responsive - I almost feel like I can fly like Marco! (HaHa - I have a looong way to go). The Loiter P from .35 to .1 helps, maybe 3 meter in low wind now,
The one thing I had that diverged from Marco was Altitude Hold P and I. Where he went down from .5 D to .2, I went to D = 1.3 and I = .01 and both sonar and baro seem very accurate and responsive. I just guessed here, it would be interesting to see what others get. The defaults were kind of soft and sloppy, maybe I should play with throttle too. The APM2 baro is amazing -usually hangs within a few inches, sometimes will move a foot or two and then recover stability.
Another note, I agree with others - I was using cheap flimsy props, finally wen to APC and It flies and sounds much better. The almost never snap at the hub like cheapies, yaw and control is much better.
Anyway, I am now anxious to go on to mission functions. It is impossible to thank Jason, Tridge, and all the other geniuses behind this. We also need to thank Marco, JLN, DaveC for thier brave testing and feedback that helped me many times.
jhahold-marcoloiterfly.param
On my standard quad, APM1, 1280 board, 2.5.3, stab is as good as always, alt-hold worked very well, even in gusty conditions.
Area too small for any other tests, but so far so good, thanks team!
Just flew… 2.5.3 on AMP2 on large Tri
1) The Yaw problem with hard nose down is gone..
2) The Yaw in general just seems to work better
3) Alt hold is fantastic… Plus or Minus 1 foot at times.. super reliable.
4) Stab is good as ever..
I tuned Loiter_P from 0.6 up to 3.0 and it just seems dumb…
It makes like one attempt to get back the it gives up and drifts 100s feet away..
Cant stay in 100x200 foot field…
Eddie Weeks
Had a crash, when i switched into Alt-hold, strong winds as you can see from the video.
But i thought i'd try the alt hold, mistake :-( As soon as i switched it rolled over and hit,
Not much damage, Bent one motor mount and broke the landing gear on one side.
http://youtu.be/PmYBD1tg5FE
I'll try to get the logs later.
Hello,
After some time struggling with 2.5 PID settings on my Y6, I managed to have a quite nice Rate flight as of today.
Firmware was updated to 2.5.3 via mission planner and I can confirm the success of altitude hold. Thanks for that. I will try Loiter later today, with flying open space.
Not always easy, but nothing is perfect.
Thanks for dev team work,
MLB
Stab PID: 5.6/0/40 (5.6 is not enough, have to try 6.0)
Rate PID: 0.145/0.01/0.005
Stab Yaw: 7/0/8
Rate Yaw: 0.145/0/0
great job guys.. simply gret
I've updated the version coming out of the mission planner to be 2.5.3 and added a new zip to the downloads area.
Patch fixes the following issues over 2.5 release:
1. reenabled automatic calculation of throttle_trim while in stabilize mode
2. enables auto trim command on Channel 7
3. throttle_trim default increased to 450
4. bug fix to clear loiter and nav controller's I term when switching out of these modes.
5. added checking that PID controller's IMAX values are always positive
6. bug fix to barometer library affecting APM1 users with temp crossing 33 deg celsius
It's this last one that's new over 2.5.2.
Looks like we've found a barometer bug during the beta testing of the 2.5.2 patch. It's there in 2.5 as well and only affects APM1 users while in Alt Hold mode (or loiter or RTL that make use of the alt hold code). If your barometer's measured temperature crosses 32 degrees celsius during a single flight it causes a wildly incorrect altitude and your copter will either fly up or down (of course you can switch back to stabilize mode to regain control).
The fix is already in trunk and I'll package it up into patch release 2.5.3 tomorrow. Sorry for the trouble.
ok, 80% tuned. Here's a current param file for my OEM 3DR hexa running 2.5.12:
Some notes:
a- I disabled auto-alt in the code, this is an arduino build and then tuned in the field. Winds were about 5mph with 7-8 gusts.
b. IF you switch on alt-hold before the vehicle absolutely hovers, it WILL jump about 6-9feet before settling back to the correct alt. 4 out of 5 times I tried reducing throttle_trim to stop the hopping, but then it will settle lower than the setpoint and level flight is not obtained (more like parabolic flight in pitch/roll). Achieving a perfect hover appears to be crucial on how alt-hold responds. If I set a really good manual hover, switch to alt-hold and no jumps, perfectly holds and flies level.
c. Appears throttle_trim is effected by battery cell count, between a 3S and 4S I had to reduce trim by 100. A 4S on 3S always jumped 3-5 feet.
d. Yes, as most mention loiter is more sensitive to the PID settlings.
e. I still had a couple of yaw jitters, but luckly in simple more, it was corrected.
Tried RTL, Waypoints, and it appears at the end of a waypoint mission it attempts to land (??? not good). Logg show land mode was activated... Anyway, all works, but need more tuning.
I'd be interested to what the 3DR folks come up on defaults for their hexa...
3dr-frame.param
-
10
-
11
-
12
-
13
-
14
of 18 Next