TFUNIPAYLOAD01 – Universal MAVLink Sensor Interface for UAV Payloads

TFUNIPAYLOAD01 is a universal interface board designed for seamless integration of custom sensors with PX4 or ArduPilot-based UAVs, especially when no dedicated driver exists for the sensor within the flight stack.

Its key advantage lies in the use of MAVLink Tunnel packets, eliminating the need to modify autopilot firmware. This solution is suited for rapid deployment and testing of new environmental or scientific sensors without the need to delve into autopilot firmware development. It provides a plug-and-play bridge between your sensor and the powerful MAVLink ecosystem.

Advantages of the MAVLink Tunnel Approach

• No firmware modification of PX4 or ArduPilot is necessary
• Standardized MAVLink interface allows inspection and logging tools (QGC, uLog, MAVSDK)
• Sensor abstraction: Only the payload firmware knows the sensor-specific protocol and interface
• Flexible and portable: The same approach works across multiple autopilot flight stacks

Hardware Overview

The TFUNIPAYLOAD01 module is based on the MLAB ATmegaTQ4401 board equipped with an ATmega1284P microcontroller. This hardware platform provides:

• 128 kB Flash and 16 kB RAM – sufficient for MAVLink message handling
• Multiple UARTs for communication with both sensors and the autopilot
• Standard MLAB form factor for mechanical and electrical compatibility with other modules

Intended Use Case

TFUNIPAYLOAD01 is intended for users developing or deploying atmospheric or scientific sensors where direct driver support in PX4/ArduPilot is not yet available or practical. Common use cases include:

• Experimental sensors under development
• Rare or proprietary measurement devices
• Quick integration of payloads without altering autopilot code

The example of this approach is the TFPM01 airborne particulate matter sensor demonstrator.

Communication Principle

Sensor data is processed on the TFUNIPAYLOAD01 module using Arduino-compatible firmware. The microcontroller encodes the measurements into MAVLink Tunnel packets, which are:

• Sent over UART to the autopilot’s telemetry port
• Logged automatically by the autopilot (if configured correctly)
• Forwarded to the Ground Control Station (GCS) with software such as QGroundControl or TF-ATMON

The TF-ATMON system is specifically optimized to receive and process such sensor data using MAVLink Tunnel messages, while enabling the TF-ATMON system to visualize and geolocate the sensor measurements in time and space. This allows users to quickly gain insight into the measured environment without requiring any sensor-specific firmware changes in PX4 or ArduPilot.

Integration Workflow

1. Connect your sensor to one of the TFUNIPAYLOAD01’s UART or analog interfaces.

2. Develop a small Arduino sketch for the ATmega1284P to:

   • Read data from the sensor
   • Format it as a MAVLink TUNNEL message
   • Send it via serial port to the autopilot
3. Wire the TFUNIPAYLOAD01 to an available TELEM port on the Pixhawk or compatible autopilot.
4. Configure autopilot parameters (see below) to enable MAVLink message logging in PX4.

PX4 Parameter Configuration

The following PX4 parameters must be set (example for TELEM2):

Parameter	Value	Description
MAV_1_CONFIG	TELEM 2	Enable MAVLink on the correct UART port
MAV_1_FORWARD	1	Enable message forwarding
MAV_1_RADIO_CTL	0	Disable radio control on that port
MAV_1_RATE	0 B/s	Unlimited rate
SER_TEL2_BAUD	57600	Baud rate (match your firmware settings)

Example Firmware

Refer to the TFUNIPAYLOAD_MINIMAL sketch for a lightweight example that sends only HEARTBEAT and TUNNEL messages:

mav.SendTunnelData(data, sizeof(data), sensor_id, 1, 1);

Function Arguments:

• data: array of bytes (max 127)
• length: length of data
• sensor_id: identifier for your custom sensor
• sysid: usually 1, or 0 for broadcast
• compid: usually 1, or 0 for broadcast

How to Verify MAVLink Message Reception

Using QGroundControl

Ensure your autopilot is connected via a telemetry radio or USB-UART bridge, not via its USB port. The following steps allow you to inspect incoming MAVLink TUNNEL messages:

1. Open QGroundControl and connect the autopilot.
2. Click the QGC logo (top left corner), navigate to Analyze Tools, then select MAVLink Inspector.
3. Look for messages of type TUNNEL. These confirm correct reception and forwarding.

Note: This requires the messages to be broadcast (sysid = 0, compid = 0), and QGC must be on a MAVLink-enabled link (not USB).

Using the PX4 Console

You can verify message reception even without broadcast using the PX4 shell:

1. Connect via mavlink_shell.py or through QGroundControl > Analyze Tools > Console.
2. Run:

listener mavlink_tunnel -n 100

This command displays the last 100 received tunnel messages.

Using Flight Logs

MAVLink TUNNEL messages are logged in PX4 .ulg logs. Although tools like Flight Review don’t visualize this type of data, you can:

• Use PlotJuggler to load and inspect message content
• Use TFUNIPAYLOAD Log Viewer Notebook to parse tunnel data programmatically

Logging and Monitoring

• Tunnel messages can be monitored using listener mavlink_tunnel on PX4
• Use QGroundControl with a radio modem to inspect TUNNEL messages in real time
• For post-processing, extract data from .ulg logs using custom tools or PlotJuggler

Limitations

• Autopilot memory is limited – avoid sending excessive data
• Max 3 MAVLink instances (incl. modem) due to PX4 firmware constraints
• Proper MAVLink stream segregation is necessary for logging vs telemetry transmission