Log Simulink Signals to DataFlash Log on ArduPilot Autopilots

Prerequisites

1. If you are new to Simulink, watch the Simulink Quick Start video.

2. If you have not done so already, complete the steps in Install UAV Toolbox Support Package for ArduPilot Autopilots in Windows.

DataFlash Log Overview

ArduPilot stores onboard telemetry data in a binary format known as DataFlash logs. These logs are written to the autopilot's onboard storage (such as an SD card or internal flash memory) and contain timestamped records of sensor readings, control outputs, navigation data, and user-defined messages.

DataFlash logs are the primary mechanism for post-flight analysis in ArduPilot. Each log entry is identified by a message type (for example, ATT for attitude, GPS for GPS data, or custom user-defined types). The logging system supports:

For more information about the DataFlash logging system, see the ArduPilot DataFlash Logs documentation.

Open Model

Open the DataFlashLog.slx model.

modelName = "DataFlashLog";
open_system(modelName)

Model Overview

The model contains two main functional areas:

Conditional Logging with Enabled Subsystem

The DataFlash Log block is placed inside an Enabled Subsystem. This design pattern allows you to conditionally log data based on an enable signal. When the enable input is true (nonzero), the subsystem executes and the DataFlash Log block writes the input signals to the onboard log. When the enable input is false (zero), no data is written.

Inside the Enabled Subsystem, the DataFlash Log block (configured with the message name LOG1, which identifies this custom log entry in the DataFlash file) accepts two input signals:

The signals connected to the DataFlash Log block are written as a custom message type in the DataFlash log file. You can configure the block to define the message name and field labels that appear when you analyze the log.

Enabling Logging When Disarmed

By default, ArduPilot writes DataFlash logs only when the vehicle is armed. For testing and development, you can log data without arming the vehicle. The model demonstrates how to temporarily enable logging while disarmed by writing the LOG_DISARMED parameter.

The lower section of the model implements this logic:

When the timestamp is between 30 and 60 seconds after boot, LOG_DISARMED is set to 1 (enabled). Outside this window, it is set to 0 (disabled).

Logging behavior

The DataFlash Log block writes custom signals into the ArduPilot DataFlash .bin log file generated on the target. You can later import that log file into MATLAB by using the ardupilotreader object, which reads ArduPilot DataFlash log files and exposes messages, parameters, and logged output for analysis.

Run the Model and Observe Results

After you complete the hardware and model configuration:

2. Click Build, Deploy & Start. After deployment, the algorithm runs autonomously on the hardware.

After you deploy the model and run it on the target, the autopilot generates a DataFlash log file that contains the custom signals logged by the DataFlash Log block. Because the block is inside an enabled subsystem, the custom signals are logged only when the enable condition is true. In this example, the model also temporarily sets LOG_DISARMED so that you can capture log data without arming the vehicle during bench testing. LOG_DISARMED enables logging before arming, while normal Copter logging begins after arming by default.

Import the downloaded .bin file into MATLAB by using ardupilotreader or the Flight Log Analyzer app to inspect the recorded signals, verify the active logging interval, and correlate the custom log data with other ArduPilot messages in the same file.

Retrieve and Analyze Log Data

Before analyzing the log data in MATLAB®, retrieve the DataFlash log (.bin) file from the ArduPilot autopilot and save it on the host computer. After retrieving the log file, use the ardupilotreader object to import and analyze the logged data.

Retrieve and Analyze Logs in MATLAB

Use the UAV Toolbox APIs to download the log file from the autopilot and read the DataFlash log data into MATLAB.

Read the DataFlash Log File

Use ardupilotreader to import the .bin file:

% logObj = ardupilotreader("flight.bin");

The ardupilotreader object reads ArduPilot DataFlash log files and provides metadata such as:

Read Logged Messages

Use readMessages to retrieve messages from the log file:

% msgTbl = readMessages(logObj);

This function returns a table containing:

Read Logged Parameters

Use readParameters to extract parameter values stored in the log. This is useful for verifying that parameters such as LOG_DISARMED were written correctly:

% paramTbl = readParameters(logObj);

Read Logged Output Messages

Use readLoggedOutput to access logged output messages from the log file. This includes custom log messages written by the DataFlash Log block:

% outTbl = readLoggedOutput(logObj);

This function returns output messages (such as custom logs written using DataFlash Log blocks).

Analyze Custom Logged Messages

In this example, the DataFlash Log block is configured to write a custom message named LOG1. Use readLoggedOutput to inspect the logged output messages and identify the recorded signal values associated with the custom log entry. Review the logged signal values and timestamps captured in the custom log entry.

% Read custom logged output messages
outTbl = readLoggedOutput(logObj);
% Display a subset of the logged data
head(outTbl)

Plot Logged Signals

Plot the logged signal values to verify that signals are recorded only while the Enabled Subsystem executes.

% Example plot workflow
plot(outTbl.Timestamp,outTbl.Sig1)
hold on
plot(outTbl.Timestamp,outTbl.Sig2)
grid on

xlabel("Time")
ylabel("Signal Value")
legend("Sig1","Sig2")
title("Custom LOG1 Signals")

The logged interval should correspond to the enable condition configured in the model. In this example, logging is enabled only during the specified time window.

Visualize Logs Using Flight Log Analyzer

You can also analyze the .bin file interactively:

Other Things to Try