The drone startup Wingtra needed data acquisition (DAQ) tools for measuring various quantities while flying. Usually, such data acquisition tools are too big for mounting on a drone – therefore I had to find ways to miniaturize them.
During my internship at Wingtra, I’ve developed three different DAQ tools measuring vibrations, temperatures and signal voltages. These tools are solely meant to improve their development process. For all three tools, I’ve got the permission by Wingtra to open source them.
For obtaining sharper images and better sensor readouts, it is critical to have low vibrations. Of course, with the spinning motors creating a lot of vibration, no easy task to improve on. With the newly developed VibrationDAQ, it is now possible to compare the performance of different damping solutions during flight.
The VibrationDAQ consists of a Raspberry Pi Zero W and Analog Devices ADcmXL3021 vibration sensor. The triaxial sensor can directly output Fast Fourier Transform (FFT) of the vibration over Serial Peripheral Interface (SPI). The main challenge was to develop a driver in C++ to interface with the vibration sensor. Additionally, a custom PCB adapter board for connecting the sensor with the Raspberry Pi had to be designed.
More details and the complete source code of the VibrationDAQ are available on GitHub.
Usually, oscilloscopes are considerably big and not meant for installation on a drone. But for some investigation, it is critical to analyze signals during flight. With the OsciDAQ, it is now possible to repeatedly record electrical signals with 30MHz+ while flying.
In a compact form factor, the OsciDAQ combines a Raspberry Pi 4 and an Analog Discovery 2 oscilloscope. With the official software, it is easily possible to set up the right trigger conditions to capture the useful part of the signal. Afterward, with a custom python script, the proprietary binary logs can be converted to CSV files. These can get analyzed with a variety of programs.
More technical details and the complete source code of the OsciDAQ are available on GitHub.
All electronics on the drone dissipate quite some heat. As this can affect sensor readouts and battery life, it is of importance to get a better understanding of which temperatures the components reach while flying.
The TempDAQ makes it now easy to monitor up to 20 temperature sensors. It can regularly log 1-wire temperature sensors, such as the digital sensor DS18S20. The python script directly stores the data as a CSV file.
More technical details and the complete source code of the TempDAQ are available on GitHub.
Project duration: Feb. – Jul. 2020