Boundary Layer Dynamic Soaring For Autonomous Aircraft

Dynamic soaring is a flight technique used by albatrosses and other large seabirds to extract energy from wind gradients in the atmospheric boundary layer over the ocean. This technique enables them to fly for extended periods without flapping their wings, in some documented cases circumnavigating the globe. This work examines the application of dynamic soaring to propulsion of small unmanned aerial vehicles (UAVs). First, the equations of motion were derived and the energy transfer mechanisms were explained for a vehicle flying in a spatially and temporally varying wind field. Next, a robust and efficient dynamic soaring trajectory optimization method that forms the foundation for the remainder of the research was outlined. This method was used to solve for optimal periodic trajectories through a number of different wind fields. It was also used to investigate UAV designs that have the ability to extract electrical energy from their environment and store it on-board, allowing operations during lulls in the wind. Vehicle speed polars were generated that show the maximum cross-country speed achievable as the a function of the wind speed and the cross-country flight direction. The method of isochrones was applied to the long-range routing of dynamic soaring vehicles across the ocean by combining vehicle speed polars with satellite based ocean wind measurements. As part of this work, a small UAV, dubbed Mariner, was designed to demonstrate autonomous boundary layer dynamic soaring over water. The objective of the UAV design problem was to minimize the required reference wind speed. Constraints were imposed on vehicle size and weight, and careful attention was paid to stability and control requirements imposed by the optimal trajectories. Sensors and sub-systems were specified to allow for accurate state estimation in close proximity to the water surface. Experimentally, a small off-the-shelf airframe was flown to aid development of Mariner's flight control hardware and software. An extended Kalman filter (EKF) was implemented for vehicle state and wind estimation and tuned through flight testing. In addition, a novel GPS tracking tag was designed and tested to improve our understanding of albatross dynamic soaring.