Evaluation of a Baseline Controller for Autonomous “Figure-8” Flights of a Morphing Geometry Quadcopter: Flight Performance

Abstract

This article describes the design, fabrication, and flight test evaluation of a morphing geometry quadcopter capable of changing its intersection angle in-flight. The experiments were conducted at the Aircraft Computational and Resource Aware Fault Tolerance (AirCRAFT) Lab, Parks College of Engineering, Aviation and Technology at Saint Louis University, St. Louis, MO. The flight test matrix included flights in a “Figure-8” trajectory in two different morphing configurations (21° and 27°), as well as the nominal geometry configuration, two different flight velocities (1.5 m/s and 2.5 m/s), two different number of waypoints, and in three planes—horizontal, inclined, and double inclined. All the experiments were conducted using standard, off-the-shelf flight controller (Pixhawk) and autopilot firmware. Simulations of the morphed geometry indicate a reduction in pitch damping (42% for 21° morphing and 57.3% for 27° morphing) and roll damping (63.5% for 21° morphing and 65% for 27° morphing). Flight tests also demonstrated that the dynamic stability in roll and pitch dynamics were reduced, but the quadcopter was still stable under morphed geometry conditions. Morphed geometry also has an effect on the flight performance—with a higher number of waypoints (30) and higher velocity (2.5 m/s), the roll dynamics performed better as compared to the lower waypoints and lower velocity condition. The yaw dynamics remained consistent through all the flight conditions, and were not significantly affected by asymmetrical morphing of the quadcopter geometry. We also determined that higher waypoint and flight velocity conditions led to a small performance improvement in tracking the desired trajectory as well.