Head Tracker runs on Arduino Nano 33 BLE and Seeeduino XIAO BLE Sense boards and connects to OpenTX and ETHOS radios.
Both wired (PPM) and wireless (Bluetooth) trainer connection modes are supported.

This project switched to Go language (TinyGo compiler) from classic Arduino C/C++ that had been archived in arduino branch.

• FrSky X-Lite Pro (OpenTX)
• FrSky X20S Tandem (EthOS)

• D2: Orientation reset pin, use button that connects this pin to GND
• D8: PPM mode activation pin, solder permanently or use positional switch to connect this pin to GND
• D10: PPM signal pin, connect to audio jack tip and GND to jack body
• SDA&SCL: I2C communication pins, connect them to 128x32 SSD1306 screen

Listed boards are perfect for a head tracker project, since they have both IMU for orientation and Bluetooth for connectivity.

XIAO BLE Sense is a tiny board with UF2 bootloader and SoftDevice (bluetooth driver) pre-flashed that makes it very easy to use. The board is relatively new and may be harder to find. Please pay attention and order "Sense" variant, it has IMU.
Recommended for users w/o experience in embedded programming.

Nano 33 BLE is a larger board from Arduino that shall be easier to source. There are two variants of this board: "Nano 33 BLE" and "Nano 33 BLE Sense" -- both will work for this project. This board has no UF2 bootloader pre-flashed and a debug probe is required to flash UF2 bootloader to it.
Recommended for advanced users who have a debug probe (JLink or CMSIS-DAP compatible) and can use it.

For a long time, Nano 33 BLE was the only board supported. It has an additional sensor, magnetometer, that head trackers usually use to eliminate yaw drift. In practice, however, magnetometer adds more problems than solves. Magnetometer is very sensitive to environment and tricky to calibrate properly. In this project we do not use magnetometer and have a good zero-configuration automatic continuous gyro calibration instead to solve the drift problem. Already after first 5 seconds the calibration is good enough to stop the drift.

As of now, Nano 33 BLE board does not provide any benefit over XIAO BLE Sense. Instead, the later board is actually easier to use, thanks to pre-flashed UF2 bootloader and smaller size. In the future, just for fun, magnetometer support can be added but only if we can make calibration automatic and transparent for the user.

You have another nRF52840-based board with IMU and want to use it? File a feature request. Better yet, make a PR directly!

Connect the board to your computer, double-tap on RST button and copy ht_xiao-ble_xxx.uf2 file to XIAO-SENSE usb drive.

First, you shall flash UF2 bootloader to the board. You only need to do this once for each new board. Then connect the board to your computer, double-tap on button and copy ht_nano-33-ble_xxx.uf2 file to NANO33BOOT usb drive.

Attach the board with flashed head tracker code to your FPV goggles.
Place your goggles on a solid surface and power the head tracker with 2 cell battery or 5v source. I use analog adaptor bay on my DJI goggles to source 5v.

On start, board shall blink continuously blue, red and green/orange leds.

• Blue led indicates Bluetooth state and blinks while not connected, it switches to solid blue upon successful connection to your radio (see below);
• Red led indicates initial gyroscope calibration, you shall wait until the red led is off before use, normally no more than several seconds;
• Green/orange led indicates health of the head tracker and shall slowly blink during normal operation.

The head tracker records initial orientation on power up, place your goggles accordingly.
Optionally, a reset orientation button can be wired to D2 and GND pins.

If you have a LED 128x32 screen added you your board (via I2C), the board's bluetooth address is displayed on it. Blinking ":" symbols indicate bluetooth connection status, like blue led. Upon start, while gyroscope is calibrating, you shall see head tracker version briefly on the screen. The version is then replaced by 3 horisonal bars, one for each axis: roll, pitch and yaw.

HeadTracker can work either in wireless (Bluetooth) or wired (PPM) mode.
Bluetooth mode is active by default.

• Flash your board with a release file
• Connect to the board with a Serial console and make note of the board address (like: 7b:f5:1e:35:de:94)
• SSD1306 LED display can be connected (via I2C) to the board; in such case board address displyed there too
• In your radio, select Trainer mode "Master/BT", wait a bit and click "[Discover]"
• Search for your board by address you noted earlier and Connect to it
• Blue led on the board shall turn on indicating successful connection
• Do not forget to configure Trainer function in your radio either on "Special Functions" screen of your model or on "Global Functions" of your radio setup.

• Activate PPM trainer output by connecting D8 and GND pins.
• Collect PPM signal from D10 pin. (Audio jack tip shall be connected to D10 and rest to GND)

• DIY-Head-Tracker
  Original DIY head tracker for Arduino Nano with separate IMU board and PPM over cable
• RC HeadTracker by Cliff
  Another version of head tracker, based on Arduino Nano 33 BLE board.
• Bluetooth Smart/BLE Crash Course
• Bluetooth low energy Characteristics, a beginner's tutorial