Heli frame, prearm Internal error 0X100000, L:602 flow_of_ctr

[hututu_duan]

I am using Copter4.2, pixhawk4, heli frame. while control the vehicle through the guided mode by a nuc computer with linux system, like the offboard mode in PX4, once i switch to the guided mode and enter the SubMode::TakeOff, an error “prearm Internal error 0X100000, L:602 flow_of_ctr” occurred.

This is an error from this line

(/home/tutu/Pictures/Screenshot from 2022-05-09 11-15-11.PNG)

Then I read the code in detail and test a lot. At last, find this is due to in Arducopter/mode.cpp/auto_takeoff_run, the SpoolState is always THROTTLE_UNLIMITED when heli. And there is no init_xy_controller before update_xy_controller. then I add the init_xy_controller in function make_safe_ground_handling and it seems worked.

Besides, I also do the same process using quadrotor, the initial error doesn’t display before and after the modification above.

It seems a problem only with heli frame, could someone tell me is there a better solution?

[rmackay9]

@hututu_duan,

Txs for the report, I’ve added it to our 4.2 issues list.

This may be one for @bnsgeyer to look into.

[bnsgeyer (Bill Geyer)]

@hututu_duan can you post a log demonstrating this issue. You will have to set the LOG_DISARMED to 1 in order to capture data for this issue. I want to make sure that I can reproduce this in our SITL.

@rmackay9 I will need help from @Leonardthall as I’m pretty sure he has been moving some position controller stuff around that may have caused this.

[andyp1per]

FWIW I see this on copter from time to time. I think we should just remove the internal error - it’s often wrong and its a pain that you then have to reboot in order to take off again.

[hututu_duan]

2022-05-07 12-23-50.log

This is the log file in real flight

[bnsgeyer (Bill Geyer)]

@rmackay9 @Leonardthall
This is a serious bug that needs to be corrected prior to releasing 4.2.0. It will cause a core dump with heli on the ground rotors turning. Very dangerous.
Here are two ways to cause it

1. guided mode takeoff
   -arm the heli in stabilize
   -Enable motor interlock and wait for spool up to complete
   -set guided mode
   -in mission planner, right click on map and select takeoff and set an altitude
   At that point the controller crashes
2. auto takeoff
   -set MIS_OPTIONS to allow arming in auto
   -set auto flight mode
   -arm the heli
   -enable motor interlock
   After the rotor spools up, the controller crashes

[rmackay9]

@bnsgeyer,

ok, thanks for looking into this and providing the method to reproduce it in SITL.

BTW, the core dump will only occur in SITL though because we treat internal errors as PANICs to make sure they’re not missed.

[hututu_duan]

in real flight the error occurred but the heli still take off with poor position control.

Looking through the code, i found it is because we run update_xy_controller without init_xy_controller.

[rmackay9]

@hututu_duan,

4.2.0-rc4 includes the fix fix for this issue. Could you give it a test to be sure you’re happy that this has been resolved?

[hututu_duan]

Tests in sitl and real flights all success. Txs.

[makeflyeasy]

@rmackay9 @hututu_duan
Testing the quadplane4.3.1 firmware today, I found this issue 0X100000.
This error also occurred when I was using version 4.1X 4.23.
It appeared during the take-off phase of the automatic route flight, and it was normal for me to take off with the remote control.

Here is my flight control log:
https://drive.google.com/file/d/1rxwDWnRHa2MVrdsmQj0ZWKQGP2lejbvU/view?usp=sharing