Previous Abstract Return to Session D2 Next Abstract

Session D2: Robotic and Indoor Navigation

Development of a Flexible Hardware and Software Platform for UAV Research
Sam Christensen, Utah State University
Location: Galleria I/II
Alternate Number 3

Unmanned aerial vehicles (UAVs) are applied in a variety of industries around the globe. This demand has created opportunities for research in the development of more sophisticated piloting algorithms. Also, the broadening application of these autonomous solutions concerns many observers. In May, a US representative introduced a bill appropriating $2,000,000 each year to “fund research on the influences of highly automated vehicles” [1]. Potential applications of UAVs range from commercial delivery drones to air taxis, to swarm-based weapon systems. Meeting market demands and demonstrating consistent operation requires a flexible and reliable research UAV platform.
Dr. David Mathias, Assistant Professor at the University of Wisconsin-La Crosse developed a low-cost drone design tailored to student researchers. He selected a 3DR Iris+, an autonomous vehicle platform capable of expansion with additional sensors [Figure 2] [2]. His platform configures an autonomous vehicle using a 3DR Iris+ ready-to-fly airframe, PixHawk flight controller and autopilot [Figure 3], and a companion computer similar to the Raspberry Pi shown in Figure 4. Although a capable platform, integration with a motion capture system and more powerful computing hardware is needed for current research. Furthermore, Guidance, Navigation, and Control software sandbox is needed to enable rapid design, implement, and test cycles. Such a system is not readily available in the open-source nor academic communities.
The proposed research project aims to fill this research need by providing well-designed, well-documented, flexible hardware and software platform for drone research. This platform will enable advanced UAV research at Utah State University and will provide a starting point for researchers at other universities and commercial entities. The overall objectives of this project are to develop a flexible research drone hardware and software platform based on the PixHawk autopilot, Robot Operating System, and Optitrack Motion Capture system. This platform must be able to support a variety of research algorithms and implementations to enable further development of autonomous testing. Documentation of the system design via a conference publication will enable further development and implementation of this platform in further research designs. In conclusion this research will publish the developed materials to the open-source UAV research community.
Plan to Meet Objectives
Design and build a flexible drone hardware platform controlled by a PixHawk/PX4 autopilot and carrying a payload consisting of an Nvidia Jetson Nano, Firefly EO camera, and time-of-flight camera.
Integrate the PX4 Extended Kalman Filter (EKF) with position measurements from the Optitrack Motion Capture System.
Develop a Guidance, Navigation, and Control (GNC) software playground which provides a simplified interface between the Robot Operating System (ROS) and the PixHawk autopilot.
Implement path follower and path manager algorithms inside the GNC playground.
Verify the correct operation of the PX4 EKF, path follower, and path manager via hardware demonstrations in the USU Autonomy Lab.
Document the system design in a conference publication and publish the open-source code in a GIT repository for other research institutions.
Present the results at the Institute of Navigation PLANS 2020 conference in Portland, Oregon.



Previous Abstract Return to Session D2 Next Abstract