Could not activate Auto Mode initially

[acsrandd1 (Greg Wilson)]

My Quadcopter: Pixhawk Clone (<1M mem), Plastic Frame, APC props., v4.0.6 copter (legacy), MP v1.3.74
.bin file: https://drive.google.com/drive/folders/1oJu4VgNgmyL_zD6jq1j20pBN_rVmGqK6

Since my last submittal I have addressed: Z axis Vib., motor alignment, and corrected the large differences in motor speeds. But I now have some additional issues that I do not understand.

I would appreciate it if someone could take a look at my .bin file and help me understand what is going on.

For this flight test I was testing Auto Mode in a more confined flying area. The problems I encountered were in the time frame of: 16:58:06 through 16:58:41:

1. EKF primary change to 1?
2. Failsafe (I am not clear what Failsafe was triggered) 16:58:10
3. Mode change failed, message: “mode change failed: Loiter requires position” 16:58:18
4. Mode change failed, message: “Auto requires position” 16:58:21
5. Then at 16:58:25 “EFK2 IMU0 is using GPS”
6. At 16:58:42 the mission “Takeoff” was executed, the quad. successfully flew the 10 waypoint mission and auto landed.

The quadcopter had been sitting for several minutes and I thought it had fully acquired the GPS sats. But from the log files it looks like there was some issue with GPS (or other issue) that created the “requires position” error message. GPS had 14 sats at the 16:58:10 time frame and through approx. 16:59:35 when sats jumped to 18. So this does not explain #3 and #4 above??

Also, the Failsafe (#2 above) at time 16:58:10, I do not understand this at all?

[amilcarlucas]

Auto mode is only available if you have GPS lock.
You can only arm and takeoff after you have GPS lock.

[acsrandd1 (Greg Wilson)]

amilcarlucas, I am aware that GPS lock is required, but the HUD shows GPS 3D Fix (see attachment).

[dkemxr (Dave)]

The Big Red EKF is telling you something. Click it and there will probably be 2 position errors also shown in red.

[acsrandd1 (Greg Wilson)]

dkemxr, the EKF error has been an on going issue that I have not been able to resolve and have not received much feedback from the forum. The EKF tends to be a transient situation, momentarily shows up in most cases and clears with in a few seconds (this screen shot was taken just prior to clearing). So, the work around if EKF does not clear is to cycle power. 90% of the time cycling power resolves the EKF issue and cycling power is required 90% of the time to ARM the Quad. Any suggestions thoughts?

[dkemxr (Dave)]

What firmware target (not in front of the PC to look at the log)? Try Pixhawk1-1m if you are not using it now.

[acsrandd1 (Greg Wilson)]

@acsrandd1

v4.0.6 copter (legacy), MP v1.3.74

Dave, loaded v4.0.6 copter (legacy), I believe this is the 1-1m version?

One more question, do I have to recalibrate mag., radio, and INS every time I upgrade to a new release?

[dkemxr (Dave)]

You have fmuv2. Try loading Pixhawk1-1M. This is from a Pixhawk with 1Mb flash:

If Mission Planner gives you trouble use QGroundControl and select Ardupilot/Chibios/Pixhawk1-1M. I’m not sure if this will fix your problem but it’s the right version for the board. When this FC was in a craft (quite awhile ago) it’s what I used.

@acsrandd1

One more question, do I have to recalibrate mag., radio, and INS every time I upgrade to a new release?

No. Only if there is large jump in versions. You only have to re-calibrate the mag if you make a hardware change in the craft.

[acsrandd1 (Greg Wilson)]

Thanks Dave, still investigating and trying to resolve.

Post note, the RED EKF is because of flags: velocity_horiz Off and pos_horiz_abs Off. Both of these flags clear in 2 to 4 min.

[amilcarlucas]

This might be temperature related. How cold was it when you calibrated, and how cold is it when you want to fly?

[acsrandd1 (Greg Wilson)]

amilcarlucas, the weather is pretty moderate, so cold would not be a factor. Good thought though.

[dkemxr (Dave)]

@acsrandd1

Post note, the RED EKF is because of flags: velocity_horiz Off and pos_horiz_abs Off. Both of these flags clear in 2 to 4 min.

They will probably go away when you see these message(s)
EKF3 IMU0 is using GPS
EKF3 IMU1 is using GPS

Depending on the FC and the version of copter you may only have 1 and it could be EKF2.

[acsrandd1 (Greg Wilson)]

Dave, I do have IMU0 and IMU1 on my Pixhawk Clone (mem<1M), yesterday I waited longer (about 5 min.) for everything to acquire and settle out, then flew 6 auto missions. Seems like some of my problems were associated with the upgrade from 4.0.5 to 4.0.6:

1. EKF primary change to 1?
2. Failsafe (Failsafe no longer showing up)
3. No longer experiencing the Mode change failure, message: “mode change failed: Loiter requires position"
4. No longer experiencing the Mode change failure, message: “Auto requires position”. I do not understand this?!

The next layer to the onion skin is the quad is cutting the waypoints (9 waypoints several yards apart) see pics pre and post flight plan/path.

Also, any ideas regarding why I have to cycle power before flight (90% of the time I have to cycle power or the Quad will not arm).

Thanks for your help!

[dkemxr (Dave)]

@acsrandd1

The next layer to the onion skin is the quad is cutting the waypoints (9 waypoints several yards apart) see pics pre and post flight plan/path.

There is a fix coming generally for this “cutting waypoints” called S-curve navigation. Some detail here:

S-Curve Progress

Hi all, I thought some people may be interested in the progress we have made with the development of S-Curve navigation. We have got the first proof of concept working and so far it is living up to all my expectations. Here is a video showing the performance of 4.0 navigation. This video highlights the problems with the current approach when we use low accelerations and high speeds. The first half of the video looks pretty good because the acceleration is only 1/3 the speed. When the acceler…

It’s not clear to me if this is in Master now. The PR for it says rebased to master but then Merging is blocked. But no matter for you unless you are running Master.

Or you can try adding a small delay at the Waypoint, increase WPNAV_ACCEL and reduce WPNAV_SPEED.

@acsrandd1

why I have to cycle power before flight (90% of the time I have to cycle power or the Quad will not arm).

Look at the messages in Mission Planner for a clue as to why it’s not arming.

[amilcarlucas]

I have a PR to make it stop cutting corners. you might want to try that until the S-curve stuff is there.

I still think your problems are temperature related. After a few minutes the temperatures stabilize and you can arm. The good news is that there are some PRs coming that will fix that as well.

So stay tuned.

PS: 4.0.5 vs 4.0.6 which one is better for you? Can you elaborate on that?

[acsrandd1 (Greg Wilson)]

Thanks much Dave, I will look into it and report back.

[acsrandd1 (Greg Wilson)]

Dave, ok here is what MP posts in the messages tab (2 conditions shown below, 1st is the initial power up, then the power cycle:

Initial Power up (cannot ARM): The msg “Hardware safety switch” and “Gyros inconsistent” continuously repeats

2/8/2021 3:42:11 PM : PreArm: Gyros inconsistent

2/8/2021 3:42:11 PM : PreArm: Hardware safety switch

2/8/2021 3:41:41 PM : PreArm: EKF2 Yaw inconsistent by 37 deg

2/8/2021 3:41:41 PM : PreArm: Hardware safety switch

2/8/2021 3:41:12 PM : PreArm: EKF2 Yaw inconsistent by 33 deg

2/8/2021 3:41:12 PM : PreArm: Hardware safety switch

2/8/2021 3:40:12 PM : PreArm: EKF2 Yaw inconsistent by 25 deg

2/8/2021 3:40:12 PM : PreArm: Hardware safety switch

2/8/2021 3:39:41 PM : PreArm: Hardware safety switch

2/8/2021 3:39:12 PM : PreArm: EKF2 Yaw inconsistent by 24 deg

2/8/2021 3:39:12 PM : PreArm: Hardware safety switch

2/8/2021 3:39:09 PM : Frame: QUAD

2/8/2021 3:39:09 PM : RCOut: PWM:1-12

2/8/2021 3:39:09 PM : ChibiOS: d4fce84e

2/8/2021 3:39:09 PM : fmuv2 00330017 31345103 36313935

2/8/2021 3:39:09 PM : ArduCopter V4.0.6 (13b6478d)

2/8/2021 3:38:44 PM : Frame: QUAD

2/8/2021 3:38:44 PM : ArduCopter V4.0.6 (13b6478d)

2/8/2021 3:38:42 PM : PreArm: Hardware safety switch

2/8/2021 3:38:39 PM : Frame: QUAD

After Cycling the Power, I can then ARM the Quad. after this list of messages posts:

2/8/2021 3:59:20 PM : PreArm: Hardware safety switch

2/8/2021 3:59:12 PM : EKF2 IMU0 is using GPS

2/8/2021 3:59:12 PM : EKF2 IMU1 is using GPS

2/8/2021 3:58:52 PM : u-blox 1 HW: 00080000 SW: 2.01 (75331)

2/8/2021 3:58:51 PM : PreArm: Hardware safety switch

2/8/2021 3:58:50 PM : GPS: u-blox 1 saving config

2/8/2021 3:58:47 PM : EKF2 IMU1 origin set

2/8/2021 3:58:47 PM : EKF2 IMU0 origin set

2/8/2021 3:58:38 PM : EKF2 IMU1 tilt alignment complete

2/8/2021 3:58:38 PM : EKF2 IMU0 tilt alignment complete

2/8/2021 3:58:37 PM : EKF2 IMU1 initial yaw alignment complete

2/8/2021 3:58:37 PM : EKF2 IMU0 initial yaw alignment complete

2/8/2021 3:58:35 PM : GPS 1: detected as u-blox at 115200 baud

2/8/2021 3:58:34 PM : Barometer 1 calibration complete

2/8/2021 3:58:33 PM : Calibrating barometer

I have conducted mag. calibration more than once in an open area using Mavlink (RF) to reduce influence of metal objects.