Image Dependent Relative Formation Navigation for Autonomous Aerial Refueling

This research tests the feasibility, accuracy, and reliability of a predictive rendering and holistic comparison algorithm with use of an optical sensor to provide relative distance and position behind a lead or tanker aircraft. Using an accurate model of a tanker, an algorithm renders image(s) for comparison with actual collected images by a camera installed on the receiver aircraft. Based on this comparison, information used to create the rendered image(s) is used to provide the relative navigation solution required for autonomous air refueling. Given enough predicted images and processing time, this approach should reliably find an accurate solution. Building on previous work, this research aims to minimize the number of required rendered images to provide a real-time navigational solution with sufficient accuracy for an auto-pilot controller installed on future Unmanned Aircraft Systems.