APM Copter 2.9.1.b has been released to the Mission Planner and is also posted in the downloads area.
2.9.1b is a maintenance release incorporating changes to 4 default parameters that have been optimized since 2.9.1 was released. There are also 4 minor bug fixes. There are no code function changes or additional features.
The changes are:
1) reduce default INS_MPU6K_FILTER to 20hz
2) reduce InertialNav Z-axis time constant to 5 (was 7)
3) increased max InertialNav accel correction to 3 m/s (was 1m/s)
4) reduce yaw_rate P default to 0.20 (was 0.25)
5) bug fix for alt_hold being passed as int16_t to get_throttle_althold_with_slew which might have caused problems if you climbed over 320m.
6) bug fix for throttle after acro flip (was being kept at min throttle if pilot switched out of ACRO mode while inverted)
7) bug fix to acro trainer to do with roll correction
8) prevent cli from being entered more than 20seconds after reboot.
Users are recommended to:
Update Mission Planner to 1.2.41 and
Update their flight code to 2.9.1b from the Firmware tab on Mission Planner. There is no need to go through a new configuration process in the Mission Planner.
Replies
I took my octo up today to work on tuning Throttle Accel P and I and had some surprise switches to RTL. I took off in acro mode and when I got to about 4 meters it seemed to stick there and I couldn't bring it down. Even with zero throttle. I wasn't in hearing rang of my computer so I didn't hear it went into RTL. I had to switch to Stable to bring it down but it would also switch into RTL from Stable. I know my Tx is ok as I just did a FPV flight with a quad at some distance. I have battery fail safe turned off so I know it wasn't that, plus I had a full set of batteries. I took it up several more times and it always did the same. After 15-30 seconds just climbing to 4-5 meters, it would switch to RTL. How can I see what triggered RTL? I can post the log if it helps.
I have noticed some posts re this but I am hoping that 2.92 addresses the issue I have noticed in the past couple releases - if I fly under with sonar under 20 feet alt hold is perfect, but as it goes above sonar range it is obviously having an issue- it will read baro, decide to go down, read sonar, go up, and it bobs up and down sometimes very fast. Flies great with just baro, sonar unit is well debugged and works fine.
Thanks to all- we are getting very close!
Here my latest video... ciao!
http://diydrones.com/profiles/blogs/arducopter-the-best-open-source...
Finally got ten minutes between showers to try the new quad again after repair. 2.9.1b
Just some loiter and alt-hold a little back and forth in the yard to check it out. Needs tuning, but on the whole, pretty pleased. With the 2650 battery, it hovers at about 45% throttle.
Quick question- My Udrones hexa C flies well using the default params with a single 5000 mAh battery and a point n' shoot camera mounted.
However when I add a second battery for extended flight time, my altitude hold no longer works properly. I can fly it in stabilize just fine (it hovers at half throttle), but when I switch on alt hold or loiter, I see sharp drops and spikes in altitude. It gets even worse when flying its route in auto- the copter frequently shoots upwards and drops downwards.
So what should I tune to fix it? I briefly tried adjusting throttle_P as well as alt_hold_P, but it only got worse. Help?
HI all, One simple question: which parameter in the Advanced Parameter list of MP controls the descend rate during RTL, immediately after RTL_LOIT_TIME has been completed? I know the final descend rate is: LAND_SPEED, but the initial descend rate is way too slow for me. I know it must be there, but I must be overlooking it.
Also, when I upgraded to 2.9.1b, my RATE_ROLL_P was reset to default (0.12), while I knew it was something like 0.14. STAB_ROLL_P is still what I set it too. Unless I'm horribly mistaken?
Don't forget turn off 'simple' mode when flying FPV.......I learned the hard way yesterday...
A couple videos of 2.9.1b on my octo. Amazing loiter. I haven't pushed the stick yet because I still working on tuning rate I and IMAX. I'm matching Rate I to Rate P but I think I will bring that up some more. It's a good starting point though.Will do the hanging something on one arm tuning tomorrow. It's very solid in the air and seems to be locked in to my roll pitch movements. I'm using APM2.5 with uBlox. I get six satellites under a faraday cage of criss crossed rebar in cement. 1/2 meter granite walls and a 1000 meter mountain in front of me. With the mytek I got zero. I'm not using the auto declination but was using before and the loiter was the same. I think the improved loiter over my smaller quads is because of it's weight. Without the camera and gimbal it's 5.2 kilos. I get some strong winds here almost daily so will be doing plenty of wind testing. I get such strong updrafts that my quad will lose yaw ability if you fly through an updraft and try to resist it going up. The throttle gets so low that it can't control the yaw and the only solution is to try and fly out of it.
https://www.youtube.com/watch?v=Z6QyCTh7EYA&feature=youtube_gdat...
https://www.youtube.com/watch?v=nUa2LI6B3qM&feature=youtube_gdat...
Just in case I get in this situation again, would switching to Acro be the solution to shut down the motors? I didn't think of that at the time with my smaller quad that went nuts indoors but with my octo that got stuck in Loiter, once I landed by dropping the throttle, I switched to Acro and I think that is when the motors shut off. If Acro gives me back full control then that is workable solution but I still think DownLeft should always disarm, even when in auto mode. Maybe it could be a user preference for those that don't want it.
I had two situations where I couldn't shut down the motors when in 2 different auto modes and I think it's potentially dangerous, especially on a larger multicopter. I wrote yesterday about a small quad that got out of control because it went into battery failsafe and tried to land. Because I had my Rate-I set to high, it bounced around trying to land, broke 3 props and almost caused worse damage. I tried to shut down the motors with a left down stick but they would never shut down. It was like it was still trying to land even though it was impossible because of my too high I term. Only solution was to grab for the battery power. I think no matter what is happening, even falling out of the sky, a left down to disarm the motors should ALWAYS work. Sort of a kill switch. I even thought a sort of auto kill switch had been implemented some versions ago so that if a copter did fall with some problem like a lost prop, it would not chew up the ground while trying to unplug the battery.
Now today with my larger Octo I had a similar situation. I took it up and put it in Loiter. BTW, the loiter is spooky to see compared to past experience. I think it only moved inches in a 5 minute loiter. I'm using ublox and the octo is pretty well tuned.
Anyway I went to switch it out of Loiter, back to stable so I could land it and it stayed in Loiter. Now that I think could have been a switching problem on my radio. I didn't have my computer on to see what mode it thought it was in but from my standpoint, it was stuck in Loiter. Then I thought what to do. I could have switched to Acro but it was dark and I didn't want to risk losing control over it so I just brought it down in Loiter by lowering the throttle. It landed very nice and then I went to shut off the motors with left down stick and it would not shut off!!! Why doesn't it disarm with left down stick like normal? I finally switched it to acro while the motors were turning at low speed and then they stopped. I carefully walked up to it and unplugged the FOUR batteries powering it.
So is this how it's suppose to be? I think if it is then it's very unsafe, especially with a powerful octo. Disarming should always be possible no matter what mode it's in right? I don't see any harm in allowing a disarming in any mode. It would never be done by accident and it would avoid a dangerous situation. It's not a good feeling to know you can't turn off the motors in any situation. If it was like my quad, that was bouncing around trying to land, it could destroy itself before you could get to it to unplug the battery.
Can someone confirm that this is what should happen or is it something unique to my setup?
My experience after installing 2.9.1.b after 2.8.1.
I did install the default values from 2.9.1b. I did the compas calibration danse, I did INS_MPU6K_FILTER to 20hz.
On Stabilize and ALT_HOLD works normal, pretty stable. I did try both INS_MPU6K_FILTER to 20hz and 0 Hz.
But on Loiter and Auto, quad behaves like crazy.
If I put in Loiter from Alt Hold in a stable flight, Loiter looks decent. But if I try to yaw in Loiter, quad starts to fly like crazy in big circles 4-5 seconds until stabilize again in Loiter.
Then I tried Auto with a mission with a fix position and 12 directions at 30 degrees, 10 seconds on each direction.
When it try to change the direction, it goes like crazy first 4-5 sec, then stabilize again. After 10 sec, issue is repeating, at each yaw, go like crazy until stabilize.
I do have Loiter radius set to 1m.
Can someone points me which PID I need to change to have Loiter more stable when try to yaw (manual or auto)?
Same quad on 2.8.1 was doing this auto yaw mission smotly, but indeed not keeping a good alt hold.
Now in 2.9.1b, Alt hold is better, but Auto with yaw correction is really agresive.
-
20
-
21
-
22
-
23
-
24
of 25 Next