Welcome to the documentation of mavlink-inspector¶
Utilities for MAVLink log inspection¶
mavlink-inspector is a package for processing and analyzing MAVLink logs as well as ArduPilot DataFlash logs. Currently, the development efforts are based on the MATLAB framework. Ports to other languages may be adopted in the future, as need arizes. Developoment efforts are targeted around the ArduPlane firmware. ArduCopter and ArduRover are not supported.
Why Use MATLAB?¶
- Other tools exist for MAVLink log parsing and analysis. Notable examples are:
- the ArduPilot LogAnalyzer toolbox, for DataFlash logs, in Python
- the MAVExplorer utility of the MAVProxy package, in Python
- Dronekit’s LogAnalyzer (Dronekit-LA), for MAVLink and DataFlash logs, in C++
On the one hand, I am not adequately familiar with Python to use it for rapid development. On the other hand, C++, while fast, it is simply too cumbersome for code prototyping. Contrarily, I am fluent in MATLAB and, to be honest, nothing can beat it for churning out working algorithms quickly.
Workflow¶
The basis of the workflow is the transcript of a .bin or .tlog log file, which usually has a .log ending. This is a simple .csv file. Each flight log is given a separate folder under the logs directory.
This is then parsed to produce the .mat data file, containing the message arrays and other helper variables.
The .mat file is finally loaded into the workspace and its contents are parsed by the various test functions.
Note
Afterwords, the message arrays are parsed to extract information and statistics. The data is not processed in one pass. Instead separate functions go over the data separately.
A Quick Example¶
You can find instructions for running the ArduPilot’s LogAnalyzer tests on an example dataset in the Examples section.
Current Progress¶
Log types¶
Initially, only ArduPlane DataFlash logs (.bin) will be supported. MAVLink logs (.tlog) support will follow.
Available tests¶
At the time, ArduPilot’s LogAnalyzer test functions are ported in. Afterwards, Dronekit-LA will be ported and after that new tests will be developed.
Requirements¶
In order to utilize mavlink-inspector you need both Windows and Linux environments, and the following software:
- Mission Planner, available on Windows.
- A Linux MATLAB installation
- Git installed in your Linux environment
Instructions for Use¶
TBA
Licence¶
The mavlink-inspector software project, which is a MAVLink log parser and processor, uses the GPLv3 license.