The latest AC2 code is posted. This is the last code before the big switchover to APM2 codebase.
What you should expect: More stability and better handling, especially if your CG is not centered.
Heres the update list from the downloads section:
Most stable flying version yet. Includes new stability patch by default and a new DCM gain and clamp value that should reduce drift due to linear acceleration in missions or long flights.
Switched rate gain calculation to use raw IMU rather than the DCM's omega. This crisps up the response a lot.
1280 has lost local flash logging, sorry, but it's just not supportable any longer.
some older tests have returned now the program memory is freed.
Remember, to use the new logging format on the 2560, you need to erase it at least once to format it.
He're the video showing just the control loop patch. The DCM patch was added later and improved the handling some more.
Jason
Replies
2.0.55. After 1-2 minutes of flight my quad begins to increasingly flip to one side and tend to crash. Feels like it opposes or even tends to return to location it's lost. Also the camera support leans to the opposite side as if IMU's internal horizon doesn't match the real horizon.
P.S. I have no compass, also changed no config due to basic X-config.
I tried 2.0.55 in my stock 3DR quad.
Loiter seems OK. There was some pulsing and log shows something on with alt. but it seems to stay in position at least as well as 2.0.49. 49 didn't do the pulsing.
I sent it off on a mission. It headed off to WP1 then flew on by to crash in a direct line through WP1 but well past it.
I haven't a good chance to look at the log, but I'll post in case it's useful. I only have a 1280 board in the quad, so there's only the tlog.
TLOG
MISSION
Over California there is a high pressure system with cold air and no winds. This makes a perfect test.
Hello to All,
Here, below, my latest working PID setup that I have used successfully with the v2.0.55 + alt_hold patch, I hope this will help you for your tuning.
The tuning of the Throttle_rate is very important to get a smooth flight during the Alt_Hold.
Below the video of the last WE tests flights with the Arducopter v2.0.54 + stability patch:
I hope that the weather here will be less stormy next WE... I shall soon post more flights videos with the v2.0.55+ ASAP.
Good tuning with your arducopter and enjoy flying...
Regards, Jean-Louis
Tested loiter with these settings and 4S 5000 nanotech and results were the same. Very aggressive over shooting just like previous video. Is this normal loiter performance or should I look for some interference? What settings should I look at in the log to see if I'm getting some wrong performance?
Weight with 4S5000 is 1732g
I tried out .55 yesterday and first with the stock PID settings on my stock 3DR frame. Only added wait is a light gopro payroll and I'm using 3S and 4S. Can't remember what I used in this. It was the first time I engaged Loiter and it took a lot to resist switching back to Stabilize. Really lurched around a lot and the wind was not that bad. Just a light breeze. I then set up my PID exactly like Jean-Louis posted earlier in this thread. It was the first time I changed from the default settings. His settings make the copter much smoother and wind seems to not effect it at all. It would be perfect for camera or fpv but not for any aggressive type of flying.
I haven't tested Loiter yet with these settings but will tonight. Loiter with the default PID was unusable.
BTW, I'm not using sonar.
Board is sitting inside soft styrofoam. It breaths well so I don't think it would effect the pressure sensor.
Ok, today I'm really negative about my copter :-( ...if anyone feels like helping or encouraging me...
I'm facing different problems with altitude hold and the magnetometer.
For alt hold, it seems to work with sonar (although pids are not good at all) but when I load the log there's no valid data (just zero). In the CLI the test works fine. Anyway, when I get barometer alt hold it gets pulsing and if I donwload the log I can see that the baro outputs zero height once in a while for no reason. With the tests on CLI it looks like it's working fine (no spikes)... obviously here you have no motor running, vibrations, etc. But why? How can I fix it?
The other problem is the magnetometer. When it points east, I get 90 as output in the CLI which is perfect then I point it south and I get around 230 which is 50 degrees off... then pointing west I get around 300 and north is fine again like east...
Third problem is that my copter lean to a random side after minutes of not even too aggressive flying. I tried playing with the pids unsuccessfully. If I land and take off again, the copter is centered again. I already soldered the filtering pads and the APM is mounted on a plastic support with a some (small) kind of vibration absorbing proprierties.
And let's keep going with troubles: I did some tests about motor redundancy (keep the copter leveled even if a motor fails) and realize that if the copter goes beyond 90 degrees of pitch, it will start flipping and not recovering. Like, even if I stop the motors, wait for the copter to stop oscillating, wait some seconds and start again the copter will move to vertical position and flip again and again. Is this normal?
The stabilize_I didn't improved the recovery speed from one motor turning off while I saw that RATE_I does a great job in this... I can share some videos if someone wants them...
Last one is yaw... it always drift slowly in one random direction... Like one full turn in 60 seconds or less.
For your convenience other problems will be discussed in another post :-P
Hehe, sorry for this but today I feel powerless on my hexa.
Hi, another question.
Can you still load this firmware out of Arduino?
Regards,
Gustav
(From a VERY windy Cape Town, grr....)
Any additional tuning (except one descriped in the manual) should be applied after burning IMU? I've uploaded 2.0.55, had 2 normal flights with no significant stability increase and one - between those two - when the copter flipped like it has not been leveled.
P.S. Thanks for fixing "Missing RC Rev param" bug:)
I am not sure to take the next step. I opened up a ESC on my quad custom build in hope to do a firmware update. It has a
C8051F310 on a Turnigy Plush 40. This Silicon Labs processor is also used with the what Turnigy calls the super brain ESC.
This hardware has not failed so far and is reliable. However I may need to purchase the ESC's from DIY drones store.
-
1
-
2
-
3
-
4
of 4 Next