Originally Posted by jreise

Many thanks for this great project, it rocks! And thanks for your help yesterday with the TLS issue.

Would it be possible to change the output behavior of the bluetooth slave in the brigde version from hold to reset? If the bluetooth connection is lost, the slave currently holds the last ppm stream, but it would make more sense in this case to reset pan/tilt to middle position to be able to continue flying safely.

Joerg

I'll vote for the "go back to centre" option.

I've had a head tracker fail with the pan turned 180 degrees and the camera pointing down.
All I could see was white foam, a VTX and a some daylight! Then it was a just case of keeping it "sunny side up" and wait for the crash.

Yes I can do that on the remote board easy enough. The polar vortex over Canada right now killed my computer last night with all the wind and power flickers... Might have to buy some parts first

I would still highly suggest everyone program their tx with an override switch still to set all chs to center jic..

Originally Posted by dlktdr

I would still highly suggest everyone program their tx with an override switch still to set all chs to center jic..

Funny enough, I'd already done that on my last head tracker (Fatshark Trinity) because it needed to 'warm up' before it would center reliably.
I don't know why I thought that we needed a feature added for this

Not everyone has the possibility to use an override switch in the RC transmitter. In my case I would feed the tracker signal into the second input of my Dragonlink tx. The Dragonlink tx sits a few meters away on a tripod as a relay station. Therefore, "go back to center" would make sense in case of a connection loss.

Joerg

Return to center on disconnect for the Bluetooth slave has been added to v1.1
Also upgraded it to the accurate within a microsecond PPM output.

Thanks, it work great!

Originally Posted by Wapata

... Or you can have the video and yell "RESEEEEEET NOOOOOOOOOW" but it should be stressing. ;-)
And well, a button on the board is as effective.

...hmmm. Maybe voice activated reset via an EasyVR voice recognition shield to the BLE . Neat thought.
Seems possible, but I tend to talk/laugh/yell at other pilots when flying in formation. It may be possible, but I'll stick with a button for now.

I am however looking for possible ways to move the center reset button down to my transmitter...possibly via simply an additional wire on the current HT cable plugged into my trainer port.

(I'm not using BluTooth)

Feature Added for Slave BT Board

1) I added reset to center from the slave board. You an use either D2 or (D3 - to match the 3d printed case that was posted)
2) You can also feed PPMinput into the remote board now on pin D9 instead of the master.

Both of these inputs made more sense to me down on this board. Keeping your head clear of any cables!

The PPM channels to be overridden are send via BT to the slave board so it knows which ones to set from the BT stream. If PPM is lost the non-overridden channels will go center. If the BT is lost PPM will pass through. If both PPM and BT lost it will set all of them to center.

I also added a watchdog timer to the connection check so if there is no data in 8s it will disconnect and try to re-connect rather than get stuck sometimes.

You will need to update the Headtracker to v0.71 & the slave to v1.2. I updated the drawings in the Wiki for connections.

If you were having an issue with the mag cal visualization it's been fixed in GUI 0.72. Otherwise no changes were made to it

I confirm that it works! Thank you for your contribution, much appreciated!

Originally Posted by dlktdr

I have the same transmitter and just gave it a try. If you have it set as a External RF (SBUS) it will still allow you to choose Trainer (Master/Bluetooth) and does connect. Then just set whatever channels from the tracker you want to send to as TR(x) in the Mixes.

ChillternFlyer, updated the case and mount is included now https://www.thingiverse.com/thing:4752289

I found a pretty bad flaw in the magnet calibration on the board.. All over a few missing spaces.. oops . Please update to 0.72 firmware. It caused the magnetic readings on X,Y,Z values to go sky high and drift badly after a calibration.

You don't need to re-calibrate in the GUI, just save your current settings, flash firmware, connect and open them again.

Nice project. Thanks for sharing..
I cannot find the ArduinoNano33BLE - PPM/PARA Slave firmware in the list.
aka;;
Note: If your using the wireless bridge method #3 to hook up. Program the slave board with the ArduinoNano33BLE - PPM/PARA Slave choice in the uploader window.

O that was supposed to say PPM/Para Master.

PARA refers to board as the Master.. but in this design it's a slave. I'll fix that photo & wiki.

How feasible would it be to add a mode so that the master unit acts as a local attitude reference
and what it outputs is slave - master values for roll/pan/tilt?
In another thread, someone wants a head tracker to aim a spotlight on
his mowing tractor. The problem of course is that IMU based HTs use an absolute reference
so if the tractor turns, the HT turns, the spotlight turns.. That doesn't work.

But someone else suggested that the signals from two IMUs (one mounted on the tractor,
the other on your head) could be compared. Just subtract the pan values of one from the
other. I couldn't help but think of this board which already has Bluetooth capabilities.
I see now that you've already got two boards communicating with each other so
we're halfway there.
Basically just need one unit to act as a reference for the other. For the tractor app,
he'd take the corrected values from the master HT and run it into a PPM->PWM converter
and from there directly to the servos aiming the spot light. Anyone else could
use this for FPV head tracking while riding in a moving vehicle, which is something
I've done before only with a mechanical head tracker.

FWIW, in this mode the master unit's reset button would act locally so you can reset to align
with the vehicle.

Yeah seems like it would be not too difficult..

I was initially planning on making the master and slave code the exact same firmware just pick master or slave from the drop down box after it's programmed. But time and all I just left it as a separate firmware file. If I did merge them the slave would have all the IMU code, calibration, etc. and this would be an really easy change.

I feel I need to spend my time getting the last two issues on this worked out first before I would tackle merging them. I still want I2C to happen in the background to free up a whole bunch of CPU time is my current focus.