Version 2.5 of the ArduCopter code is now available in the AP Mission Planner and in the downloads area!
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!
Nice :D welcome to the sky
Hi all, I'd like to show you how my hexa flies :-)
Guys I am really getting frustrated these days with tuning. And honestly, I'm about ready to go buy a skywalker and set that up instead. My copter used to fly like the above video from the above previous post or better. My hexacopter is running the latest version from the mission planner. I am having a few different issues and would love some opinions on them:
1. My aileron and elevator channels responds abnormally with it being very numb around the center but when it does move it is very abrupt. It takes quite a bit of stick deflection to get it to pitch forward and then when it does it is extremely abrupt resulting in massive stick deflection in the opposite direction to try and keep it from rocketing away. I had upped the stabilize_P value from 4.5 to 4.9 and then 5.0 and it seemed things got a bit better (still no oscillations). 4.0 was completely unstable. I moved the Stab_D value down to .020 as I read it might have worked on another similar setup and I think it helped. I'm honestly tuning blindly. Making random changes it seems and hoping for the best. I've read all of the tuning guides but nothing really seems to be working. All the motors are calibrated correctly.
I tried playing with the dampening values but could not really get them to work. I'd rather set my expo values in my radio personally if that is what it refers to.
2. My sonar was working half decently and tests fine in the terminal. When I switch to alt hold while flying ,with my sonar enabled, the copter now wants to slowly rise or fall without any signs of stopping.Is this because the copter is rising or falling when I engage it?
3. I'd like you guys to see the below video to see my setup. I was flying the below setup minus the landing gear, floating camera plate and fpv gear very well around the 2.0.46-2.0.49 time period. I was out of the game for a while while I was planning a rebuild after a crash and did not have time to commit to fixing it. Now that I have reloaded and tested everything, I am having just utter shit flight qualities. It seems like the copter is always wanting to drift and it seems like my compass is always adding on about 5-10 degrees to the left after I stop yawing with the stick.
So I am wondering if the landing gear is creating any major issues with the sensors from a pendulum effect? Does that have any bearing or should I be compensating in the tuning values somehow? I would mount the batteries above the APM if it helped.
I would love to help, but here's the problem. You have changed everything and had a crash. You say it flew fine before you rebuilt on older code, but now it flies bad and your assumption here is that the code is the problem. I respectfully disagree as 2.5.5 fixed just about all the issues I had, it workes fine on my hexa and quad.
By your comments about it wanting to yaw and the compass seems to be pulling is likely is your battery wiring is near the compass and throwing it off. Again, stuff you likley messed with in the rebuild, Try twisting any power wires, and moving them as far away from the APM as possible. Simply turning the battery around in the mount so when you plug in the leads it goes another way around can make a difference here. Again, if this is happeing, no amount of tuning will matter because you basically glitch the compass and the slow auto correction to the gyros is happpening in the code.
Hi Grips. Please post your settings and what motors and battery you're using. My hexa is a bit smaller than yours, but I'm using much smaller STAB P values and I'm happy with it. I'm in version 2.5.4 but don't think there's much difference regarding stabilization laws. I'm using STAB P of 3.2 STAB_D 0.090 and RATE P of 0.100, RATE_I of 0.010 and RATE_D of 0.001 . Of course our settings are different but you may want to start with this values and see if it gets better or worse. Regarding those stick strange beahaviours, are you sure you don't have exponential curves on the channels ? Try them linear, also you might want to check the I terms, or set them to 0 just to give it a try. Avoid metallic objects near the compass (5 / 10cm) . Also if you're using a switched uBEC, put it the far away as you can from your RX since it may cause interference. If nothing this works, try yo re calibrate your ESCs one by one. Also, in order to level the hexa, what I do is use a bubble level to make it as levelled as possible (measure from top of motor to top of motor across using small thing underneath the landing gear to level it and connected to the mission planner, hit the "level" button. I noticed that doing this, in the air, without wind, my hexa has almost 0 drift :)
Don't give up !
Then, don't bash the code if it doesn't perform on your own, unique design.
Remember, the code is written, and tested, on factory standard platforms.
Personally, I'm happy with that.
It seems to work there, the videos prove it :-)
If you want large, well, there are "off the shelve" products out there.
Just add another zero or two to the price, and good luck with support and spares.
As for Fixed Wing, go for it !
I also have a ST Models Discovery fitted with an APM1, from day one it performed beautifully :-)
Can you post your dataflash logs? At least for #2 (altitude hold) it should be pretty clear if it's the landing gear that's interfering with the sonar. There's a CTUN message that appears in the log and it has "Sonar Alt" and "Baro Alt". If the two are wildly different then there's likely a problem. Or if there are lots of spikes in the sonar message is a command issue people face (usually EMI is the cause).
Also the poor roll/pitch control should show up in the ATT messages by comparing the "Roll IN" to the "Roll". Of course that just makes the problem visible, not an immediate solution but it should show us something.
For #3..there is a "yaw bounce" at the moment in which the yaw overshoots. I was looking at this over the weekend and I have a partial fix but it's tied up with a bigger change..
Randy. I have attached two screen shots of the roll vs roll-in and the baro-alt vs sonar-alt. Looks like some discrepancies in the altitude one for sure.
Right now my values are as follows
ok, your sonar is definitely bad. you can see it's constantly showing 20cm with occasional spikes. It's hard to say what's wrong with the sonar, it could be something blocking it's beam or a wiring issues. It's slightly possible (but unlikely) that there's an issue with the ADC on your board (I guess you have an APM1).
The best thing (as you might have guessed) is to try a bunch of tests to identify where the problem is (apm or sonar). I guess removing the landing gear and testing the sonar using the CLI would be the first thing (go to the MP's terminal screen, then type "test", "sonar") and see if when you wave your quad around the distances change.
If that doesn't work then try disconnecting the signal line from the sonar and connect to a volt meter (hopefully you have one) and see if the voltage changes as you point the sonar at various targets at different distances. if the voltage doesn't change then the problem is likely the sonar is broken (althought this is very rare) or the power/ground voltage to the sonar is messed up. By the way, if your hearing is good you can actually hearing the click-click-click of the sonar.
If the volt read with the voltmeter does change then it's likely a problem with the APM. I'd then try applying a ground, then a voltage (i.e. GND or 5v from some other part of the APM) to the signal pin that the sonar is normally attached to and make sure the CLI test shows 20cm, then 700cm.
Anway, thats's probably enough steps to get started..
Ah, your Roll In vs Roll looks ok really...basically the green line is following the red line..the slight delay is totally expected as it takes a moment for the commands from the APM to get to the engines etc..
I'll take those suggestions and try them later today to see if it helps. I actually do hear the clicking of the sonar so I guess that means it is working. Perhaps it is off centre a pinch and the landing gear leg is messing with the values.
What do you think of mounting the sonar out under a motor to get it away from the gear? I presumed that it would be a bad idea due to EMI from the ESC to motor wires.
I've got a large hex that I'm having trouble tuning. I've got rate_P at 0.04 and tried stabilize_P at 3.0 and 3.5 (couldn't tell a difference). I can't fly as much as I like, so my testing opportunities are few.
Is there an actual guide to tuning somewhere? I can find a sentence here or a snippet there, but no actual guide. There is a page of instructions on the wiki, but it doesn't seem to be all that helpful unless you've got a normal sized quad.
Besides the controllability issues, it seems to be confused about what level is. In this video clip, you can see about 10 seconds in and again about 2:15 in where the hex pitches back and flies toward me (I'm to the right of the camera) it's doing so with the stick neutral. I've ran the "level" calibration multiple times. Any clues?
log files attached below; sorry, forgot to get the params file before I left for the weekend.
here's another clip of it lifting 8lb of payload, for a GVW of just over 20lb / 9kg:
Jason, just going on the video it looks pretty stable to me. Likely its flying in ground wash affect if your below a meter which might be giving it the instability you see. Try some test flights above 1.5 to 2 meters. By the way, how heavy is it without pay load? I also see you have the batteries on the arms which might need different PID settings for good stability?