Flight Test Data Acquisition

My senior design project at Florida Tech was to design and implement a new data acquisition system in the flight test engineering aircraft. The goal was to modernize the data collection on the aircraft from outdated "manual" methods to automated digital data gathering. This would improve the quality of data gathered, improve the amount of data gathered, and allow students to become familiar with modern flight test techniques.

The project was broken into a number of different subsystems including flight control position, flight control force measurement, laser altimeter, angle of attack and angle of sideslip measurement, Garmin G5 data integration, and data collection and processing.

For this project I was mainly responsible for the later of the systems with data collection and processing. We wanted this system to be as easy to use and maintain as possible, so our goal was to use off the shelf components which were well documented and used throughout industry and academia. To meet these requirements, we decided on using a raspberry pi as our main processing component with a specialty board called a Pi Plate to act as the DAQ. This also allowed us to write all of the code in python which made it easy to use, learn, and edit. Lastly running the system off of raspberry pi also made it very convenient to display the data in flight. Using the remote desktop software VNC would allow users on iPad or laptops to access the pi in the aircraft and see a GUI presenting the data. This also allows the users to interact with the system in real time on the aircraft.

As part of this project, I wrote most of the code for processing the incoming data both using the DAQ plate and serial buses for information such as the angle of attack sensor (preprocessed on an Arduino) and the Garmin G5. The software ran on a loop at 10 hz to refresh data on the screen. The data itself was collected using multiple threads (one for each system) which then updated a data structure containing the most up to date data. This allowed us to continue running the software even if there was an issue collecting data from any single system or any of the threads crashed or locked up. It also allowed the software to run with some systems not connected.

As part of my work on the data system I also wrote calibration tests for each of the sensors using MATLAB. These were designed to output certain calibration coefficients from a ground test that would be used in the data processing software.

Samples of the python code can be found below.