Evaluation of an Autonomous Soaring Algorithm for Survey Missions in Simulation Environment

An algorithm for powered, fixed-wing unmanned aircraft systems to seek and navigate thermal updrafts is proposed. The algorithm consists of an outer and an inner loop. The outer loop selects between a powered cruise, gliding or soaring flight mode, while the inner loop models the low-level three-axis and motor control provided by an autopilot during flight. The soaring strategy is based on a tree architecture similar to the decision-making process of glider pilots. The algorithm does not require complex and resource-intensive computations and can be implemented in popular flight controllers, such as ArduPilot. The algorithm is tested using a simulation environment based on a three-degrees-of-freedom point-mass aircraft model and a thermal model.