Copter-4.7.1-beta1 released for beta testing!

Copter-4.7.1-beta1 has been released for beta testing and can be downloaded using Mission Planner, QGC or directly downloaded from firmware.ardupilot.org.

The list of changes vs 4.7.0 are in the ReleaseNotes and copied below.

  1. Board specific changes
  • ARKV6S board added (@dakejahl, PR:33157)
  • CORVON_V5 board added (@holydust, PR:33593)
  • DAKEFPV boards add support for LSM6DSV and BMI270 IMUs (@engineer-dakefpv, PR:33984)
  • FlywooF405S-AIO and -HD boards support SPL06 barometer (@mbuzdalov, PR:33779)
  • KakuteH7-Wing boards support LSM6DSV IMU (@zebulon-86, PR:33761)
  • ORBITH743v2 board added (@orbittechnologyx, PR:33582)
  1. Plane specific changes
  • Q_P_JERK_NE/D renamed to Q_P_NE/D_JERK (@rmackay9, PR:33843)
  • Quadplane: In transition only set TECS throttle min in auto throttle modes (@IamPete1, PR:33972)
  1. Copter specific changes
  • Auto mode climbing Arc waypoint speed limits and velocity/acceleration fixed (@lthall, PR:33833)
  • Auto mode fix to continue past RTL if rtl-final-descent non-zero (@rmackay9, PR:33797)
  • Circle mode entry fixed when RC input present (@peterbarker, PR:33525)
  • PSC_JERK_NE/D renamed to PSC_NE/D_JERK (@rmackay9, PR:33843)
  • Simple mode fixed (@rmackay9, PR:33947)
  1. Library/Driver changes
  • AC_WPNAV: Use of uninitialised values avoided in m/cm conversion code (@peterbarker, PR:33715)
  • AP_Filesystem: Files without an End-Of-Line character get correctly read (@peterbarker, PR:33744)
  • AP_FlashStorage: Fix incorrect value of current_sector during init (@Juergen-Fahlbusch, PR:33640)
  • AP_Follow: Fix parameter ranges to include their defaults (@lthall, PR:33928)
  • AP_GPS: Unicore baseline correction in vertical component of Unicore reported baseline (@lthall, PR:33730)
  • AP_Mount: Fix AlexMos yaw angle scaling (@borischernov, PR:33785)
  • AP_NavEKF: Log per-core source set in XKFS (@mykhailopylyp, PR:33904)
  • AP_OSD: use Betaflight SYM_TOTAL_DISTANCE for DisplayPort distance icon (@coreyfarley, PR:33908)
  • AP_RangeFinder: TFMiniPlus: Small delay added to improve read reliability (@IamPete1, PR:33709)
  • ChibiOS: fix overdrive disabled on H7 boards clocked above 400MHz (@andyp1per, PR:33824)
  • GCS_MAVLink: Correction in diagnostics printing (@peterbarker, PR:33800)
  • IST8310 compass gets probed only if the driver is enabled (@peterbarker, PR:32696)
  • Topotek gimbal network connection handling is corrected (@peterbarker, PR:33803)
  1. Tool changes
  • build_options.py: Wrong dependecy for MAVLINK_SIGNING has been removed (@peterbarker, PR:33798)

We would like to get the simple mode fixes out quickly so we are only planning for a 1 week beta period this time so any and all feedback is greatly appreciated!

Did about 1h of tuning and flying today, and see no problems except one small thing - I have RC function set to Mount Roll/Pitch Lock (185) and seems it’s status is not initialized on startup, but I have to switch between channel values to get correct lock mode.

On startup channel is set to HIGH, but only when I switch to MID or LOW and back to HIGH after takeoff it works correctly. But I think that was also behavior in 4.7.0

EDIT: Forgot to mention, that gimbal used is Caddx GM2.

Hi @Adam_Borowski,

Txs very much for the testing and feedback! I’ve added it to the 4.7 issues list. The fix is quite easy as long as we’re sure that we want it initialised and are sure it’s fine when no RC is attached.

FYI @hwurzburg

No RC - no problem, but I would expect at least when I arm rc function state is applied according to channel value. That works without any issues for option Range Finder Enable (10). Channel is HIGH on startup/arm and rangefinder is enabled without switching channel to LOW and back to HIGH.

In general that is just a minor issue, but a bit annoying (at least for me :slight_smile: )

This has uncovered a long standing bug with aux switch initialization code…if a function has a backend, then the init occurs before the backend is initialized and gets a null ptr which makes the sw function init silently fail… many modules have this issue… I will work a PR to fix it globally rather than just patch this one

PR 34166 pushed for this bug

Has 4.7.1 been released?

Hi @Adam_Borowski,

Yes it started appearing late yesterday and it’s all built now. I’ll publish the announcement shortly!

We have so many autopilots now that the build time is really long and variable.