Just posted this version which has an increased performance DCM implementation. This allows the main loop to run 50Hz faster without increasing the CPU load. I've flown this for 30 minutes and it's an improvement in every respect.
Circle mode now seems to be working just fine. This should be the last beta! enjoy testing!
Jason
Here is the Link to a download:
Replies
Hello,
Yesterday I flew the 2.0.42, and looked in the code.In System.pde I saw that in STABILIZE and ALT_HOLD, Yaw was computed according to the navigation parameter.
I would have expected to have YAW_ACRO for these 2 mode, but that is jsut a personnal expectation I don't know if others share this.
During the flight, I used both Channel 6 & 7 to tune Rate_P and Rate_I. Then when it was OK in ACRO, I switche to Stable and Tune STAB_P and STAB_I.
But, though the quad is stable in ACRO, as soon as STABLE mode is active, it started fast small oscilation, when Go back to ACRO they disappeared.
Would it be requiered to have different RATE_P & RATE_I for ACCRO and for STABLE mode?
In the regulaion loop, may the STABLE regulation interfere with RATE regulation?
Does anyonelse experienced this behaviour when switching to STABLE?
Thank you
I ended up ordering a new IMU, but my IMU that had weird forward pitch drift issues is flyable now with extremely high P for pitch and roll. It jitters a little bit but atleast alt-hold works very well. Loiter actually worked for the first time for me, so well in fact I left it in the air for a couple minutes while I went inside and grabbed a drink! I think the motors I am using are throwing it off, luckily I have a set of 6 pulso 2820/14's to run on 5s with 12 inch props which I think will run beautifully smooth. I just need to find a proper frame for it!
Jason I added the following on Parameters.h. Is it ok? Do I need to change others things? I just realize if we are to utilize the "low_batt" we need to set this accurately and it will be troublesome to keep on changing it via arduino.
k_param_low_voltage,
AP_Float low_voltage;
low_voltage (LOW_VOLTAGE, k_param_low_voltage, PSTR("LOW_VOLT")),
Hope you can check it or add it if possible. Thanks.
Dude!!!! Your code ROCKS!!! I can fly my quad 4 inches above my lap on alt hold! 10 out of 10!!!!!!!
Hi all, tried my first test flight today with 2.0.42, stabilize was good as usual, although it did need trimming and for the first time loiter wasn't a scary experience, I was comfortable to leave it in loiter mode without having to jump for my channel 5 switch thinking it is going to crash. Its ability to hold altitude was much improved although it did drift around in about a 10m circle, my PID's were the same as previously posted with NAV WP pid P upped to 2.5, what do I need to alter to help it hold a tighter circle. P.S. it was pretty windy today so going to test again tomorrow. Tried the kml in google earth but it just shows loads of dots. Log posted below
2011-09-14 08-02 1.log
First hand test today, after two monthes of preparation :=)
I loaded 2.0.42, did a precision leveling, checked all that could be checked.
I'm using a modified droidworx heavy load frame, about 3.7 kg actually. Hexa X setup, about 10 Kg total thrust.
I tested first rising slowly the throttle, but it wanted to flip on the ground. So i stopped this, as i think it is certainely not the right way to test. The gound was not level at all, perhaps this has an influence ?
So i decided to make a hand test. I was able to rise the throttle and get a good hovering in stabilize mode, just keeping the frame between 4 fingers, to avoid loosing the machine.
So all seems ok except the slow throttle rise on the ground where it has a tendancy to flip on the ground. Is it normal ? Should i try a real fly or do i need to correct this before ?
I notified that my throttle stick is only 1/3 of the total range for hovering. This seems normal as the motors have been calculated for about 1.5 Kg of camera / video transmitter load (not fitted actually). Could it be a problem to not have the throttle stick centered for hovering ?
Thanks to all developpers and community for all the great work. I was really pleased to see my first multicopter machine hovering almost perfectly at the first try.
Next tests later this week i hope with a real flight. The weather seems perfect.
Just tried 2.0.42 on my Quad in Stabilize, Alt_Hold and Loiter modes. Stabilize is very good but has been for a few versions now so I noticed no difference there. Alt_Hold is a little better: I had it hovering just above head height and when I stood under the Quad it up it went by about 1.7m and stayed there until I moved away! Loiter kept within about a 2 or 3m circle until the wind blew then it took a while to come back, overshooting in the process. I'm sure I can improve that with a bit of tuning.
I still get random motor pulsing in Alt_Hold and Loiter but it is not too bad and does, at least, come back to the chosen height whereas before every pulse took it higher and it never came down!
Richard
Its not broken though, it works on .38 fine, they both do.
I would RMA it but it seems like a fixable issue seeing as I can get perfect results from one without arming the motors and the other slowly drifts on one axis. What changes with the IMU when arming is done? Maybe theres a way to fix it?
Just came back from testing this version. Even though it was a little windy I had the first successful loiter experience, ever. This is really a big step from the previous versions. Tomorrow I will test waypoint and guided as this is the first time I feel confident to do so. Thanks for the great work!
Tested just before sunset but here is a little video:
http://vimeo.com/29005790
Greetings
Niklas
-
1
-
2
-
3
-
4
-
5
of 5 Next