Harnessing the Power of the Ardupilot and Pixhawk for UAS Atmospheric Research: An Integrative Approach

There is a growing interest in the use of unmanned aircraft systems (UAS) for weather research and atmospheric monitoring. Correspondingly we are witnessing rapid development in autopilot capabilities, control software, battery technology, and the airframes themselves. However, to effectively exploit these advances and fill the gap between state of the art UAS engineering and meteorological research, we must continue making efforts to further improve upon and integrate atmospheric sensors into this development flow. To address this challenge we have adopted an open-source and open-hardware approach to accomplish the integrated tasks of managing flight control of the unmanned aircraft (UA), data acquisition from our atmospheric sensors, and establishing communication with a ground station. Moreover, we are using these tools to develop a means to allow data sharing and visualization over the Internet. The core software package being used in the project is the open-source and community-driven autopilot package called Ardupilot. The Arudpilot package can be operated "as is" or readily modified by researchers according to their particular application needs. In our research program, the Ardupilot package is run on the popular Pixhawk open-hardware microprocessor autopilot unit, which can be integrated into a variety of land, air, and water-based autonomous vehicles. Moreover, the Pixhawk hardware is compatible with many common communication protocols (I2C, SPI, CAN, Serial) used by many commercially available sensors. Communication between the UA and the ground station is achieved using the open-source Mavlink protocol, which is a robust and reliable way of streaming data in real-time. Distribution of data over the Internet is achieved using the services provided by the Microsoft Azure portal, from which the data can be pulled for further analysis and post-processing. In this presentation we provide a brief overview of the hardware and software components being used for flight control; data acquisition and processing; and communication, both between the UAV and the ground station and the ground station and the Internet. We then provide examples of how these elements can be integrated into UAS to provide adaptive atmospheric sampling strategies, which are relatively platform and sensor agnostic. We plan to continue pursuing an open-hardware and open-source approach in our UAS atmospheric research and welcome the opportunity to partner and collaborate with others.