Copter-4.1.3 has been released as the official version for multicopters and traditional helicopters. The changes vs 4.1.2 can be seen in the Release Notes and are also copied below.
a) CUAV-X7 servo voltage detection support
b) Support for IIM-42652, ICM-40605 and ICM-20608-D IMUs
Bug fixes a) Autotune twitches at no more than ATC_RATE_R/P/Y_MAX param value b) Debug pins disabled by default to prevent rare inflight reset due to electrostatic discharge
c) EKF3 reset causing bad accel biases fixed (never reported on a user vehicle)
d) Main loop delay fix for boards with 16 bit timers (affects KakuteF4, MatekH743, MatekF405, MatekF765, SpeedybeeF4)
e) MOT_MIX_MAX constrained between 0.1 and 4.0 (would previously reset to 0.5 if set too high or low)
f) Polygon Fence upload fix when replacing fence with one that has fewer points
g) RC protocol detection fix that forced PH4-mini users to powerup autopilot before transmitter
h) SmartAudio high CPU load fix (previously it could starve other threads of CPU)
i) TradHeli fix for missions which continue after a Land command
Perhaps the two most important fixes are 2a and 2b (highlighted above). 2a fixes and issue that caused a crash on a large quadplane and could theoretically at least cause the same problem on a large multicopter.
2b resolves an issue found where a CubeOrange on a Kore carrier board reset during bench testing (and the same thing could happen in flight).
Thanks very much to the beta testers and other developers who contributed to this release and Happy New Year!
I’m maintaining my own Copter-4.1 branch with some commits at the tip for SkyViper Journey. @rmackay9 it would be great if you could pull it to upstream as it fixes many bugs with this quad that I think the community will benefit from. There are even commits from @tridge merged from the stock firmware that were never included in AP.
Please review the commit messages for details.
Most of the fixes apply to master as well as older versions too, and the bugs are present since 3.6.
I believe some of the patches might break things for 2450GPS, but I cannot test as I do not have one.
Let me know if I can help. Maybe we should open another thread for this?
Re the main or auxiliary pins startup behaviour it might be good to raise an issue. If this is the issue I think it is then I’ve investigated this at least twice over the years but we have not had a good way to resolve it perhaps because the bootloader does not have access to parameter values.
Thanks Randy - I have a faint memory that the current behavior is deliberate for some purpose. In lieu of an option to configure the behavior, I suppose it might make sense to poll the users population to see which option is more favorable.
For those of us that use the AUX port relay to trigger a camera - if the camera is taking still images, it only causes an extra photo to be captured. For people like me who capture video - the present behavior causes video capture to start. I either ignore it and crop off the unwanted video in post processing - or click the camera switch on my transmitter to shut off the video.
It may be that most camera users control their cameras with devices like the Seagull or AirPixel Commander - which may eliminate the problem. But for users like me who don’t use a device like this - the problem remains.
Certainly in the “nice to have” category if it can be fixed.
@swallabat do you run 4.1.3 on your 2450GPS?
I’m not in the UK but you can test my build and report back if you encounter any issues, especially with controller buttons. arducopter-4.1.3-v2450.zip (794.2 KB)