Version 2.4 of the ArduCopter code is now available in the AP Mission Planner and in the downloads area. Although not as big a change as the 2.3 release, it still includes a respectable number of enhancements and bug fixes.
Enhancements:
- massive reduction in memory use (>2k saved!) due to Parameters change and APM2 baro lib fix. TradHeli code can again advance in hand with the quad code
- improved throttle range in stabilize mode. You may find that it takes slightly higher throttle than previously during hover but hopefully you'll find that you have more useable range of the throttle stick as well.
- added TUNE_LOW and TUNE_HIGH for CH6 Tuning (I.e. to set tuning range between 2 and 6 set TUNE_LOW to 2000, TUNE_HIGH to 6000)
- new ACRO_P gain for those who like to fly in ACRO (aka Rate) mode
- Loiter improvements including scaling of X speed
- Easier to use motor testing. see the videos for quad and octa-x-coaxial.
Bug fixes:
- remove unnecessary block for next GPS value before saving home location
- remove wobble on take-off
- turn off camera pitch control (which uses channel 6) when using channel 6 for tuning
- reduced rate of decent when landing using sonar
- Add support for CH_10 and CH_11 for APM2 users
- fixed Log_Read_Control_Tuning so that it doesn't read too many bytes
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.
Thanks go to the numerous contributors including users and their detailed bug reports, developers and testers. Hopefully all together this will add up to a nice smooth release!
As per usual, please post your comments, issues in this discussion. For enhancement requests for future versions, feel free to add them to the issues list. Note: you can "star" an issue to receive emails when someone comments on the item. On the dev side it helps us because we can get an idea as to which feature requests are the most popular by sorting by the number of people how have starred each issue.
Replies
Just flew my hexa (+ frame). I also have the yaw issue.
First flew with 2.4.1 from a few weeks ago (before the dcm fixes). Flew in stabilise as usual.
Then loaded the latest git (94c14cef91ab) and I see the yaw issue. On liftoff the copter would yaw right about 30°. Also when suddenly increasing throttle the copter will yaw right about 15-20° and correct when it stops climbing.
Another thing I noticed : with 2.4.1 descends were pretty good with little wobbling. With the latest git descends are much more wobbly.
The latest git also seemed to fly more aggressive.
All flights with standerd pid's.
Here a video of a test flight done with the firmware ArduCopter v2.4.2 official GIT (version released on March 11,2012).
Frame: Quadcopter X F450 DJI Flame Wheel with ArduPilot Mega v2.0
LOITER mode (GPS position hold) tested with a 13 Km/h wind blowing from the right and gusting 20 km/h
In spite of the wind blowing from the right and the gusting conditions, the quadcopter is able to maintain its GPS position.
- There is no drift observed on roll, pitch, yaw axis after a full flight of about 10 minutes (Thanks to Tridge...)
- The altitude hold and the throttle management is now correct and the quadcopter is able to maintain its baro altitude without any bumps (good work Jason...)
- I have also tested fast speed translations and the angle of pitch is maintained and remains stable.
The tested version is an original version currently on the official GIT and not a modified version.
I really enjoy to fly with this version.
Below my tested PID
Regards, Jean-Louis
Would it be possible to have multiwii compiled for working with APM 2.0?
Of course a version without GPS, etc.
It very dificult to change all the time from one board to the other.
And for us using only stabilized right now might be a good way to keep us on the air....
Hope nobody miss understand or get pissed off because of this message.
It would be just another possibility on the same HW.
Regards,
Pablo
Latest GIT = compass wrong of some like 130°/150° in my country, declination is the same... does anyone can confirm?
I've removed from here "my flyable version" at the request of Randy, but at least with that you could fly... :-)
Don't fly with the latest GIT!
Re: version AC_2.4.1_c98f6cd2cfa2
Had two brief cases of the "leans" AGAIN this morning, in the same flight and shortly after the first take-off, strangely it seems to happen always with the first flight of the day (exactly the same yesterday morning), but after a few take-offs and landings it disappears and the rest of the flight is completely normal with no re-occurrence of the "lean".
Yaw drift is also still an issue, doesn't seem to always hold direction correctly, never happened before with previous versions.
Eve with the GPS lock in on3D and If I am flying in stabilize, I assume it wont hold position unless it is locked be loiter or other mode setting correct? So if there is any wind then the wind can easily move the copter around and we have to compensate it with stick input?.
Hi Folks,
yesterday i had a quite weird experience.
Got a Homemade Octo with APM 1 and 2.41 firmware.As i wrote before, i´m using a cam gimbal and had sort of brown outs in lack of power.So i got a 5 amps ubec, which i fitted to my frame.Nice flying,quite stable.
But yesterday, out of a sudden all motors cut of in flight and it dropped down like a stone.
When i reached the Octo, about 20 secs later,i accidently touched the throttle stick,all motors working.......
I already checked all Cables,no problems.
After checking the inflight video,no abnormal sounds ,just silence.
So i ain´t got the fainthest clue what happend.
Any ideas ???
For your future test with latest git i've posted here try to put this line in the config and give me your impression:
// default RC speed in Hz if INSTANT_PWM is not used
#define RC_FAST_SPEED 490
I leave here the parameters of my little quad, in my opinion are a good starting point, especially for the sections STAB / RATE / LOITER, with default parameters have never been able to get a good loiter, but with these yes, my videos speak for themselves.
Marco
2.5.0_Quaddino.param
DCM fix available on the master repository:
We need your reports! What we want to avoid is to release a version that has problems.
The latest code changes by Andrew Tridgell (epic job, thanks mate) are now available on the official "master branch" of ArduCopter GIT, is virtually the same version I used in my last video, is the same that you may download now by using the GIT.
There are still a few things before making it official, but for me it's definitely usable and more secure than the current release available (imho), I have over 70 flights with this with no drift or dangerous situation.
Are very welcome your test, especially for the "yaw dancer issue", we want to see if it comes in all configurations.
Quaternion is very experimental, so avoid using it.
The "Throttle Hold is dynamic", so do not worry if you see its value change in the planner, it is normal.
Important for me: before takeoff put the "STAB_I" to "0", now the default is "0.1" and is wrong for the real concept of "PIDT1".
If you decide to test it please, do it only if you have the opportunity to read the telemetry to the ground, and before you actually fly along up or hovering two feet off the ground and make sure the quad is responding correctly to your stick input, take a full flight at that altitude, moving quickly but keeping the quad at that height.
If you have questions or need information please write here too, are all day at the airfield with the notebook.
What is still missing:
- the routine that automates the "declination" parameter (by Adam)
- the new autolanding (by Jean-Louis)
- rtl with first the "safe altitude" and the "autoapproach" option
- the yaw fix
- better management of the dynamic throttle hold
- the quaternion compass auto calibration, or manual calibration (better, imho)
I remind you that this version is not an official release, so use it at your own risk!
Check the correct movements of the hud in the planner before takeoff...
Enjoy!
Marco -=- AC Dev Team
EDIT: attachment removed due need a lot of fix... (at the request of Randy)
-
1
-
2
-
3
-
4
-
5
of 11 Next