On the scalability of experimentally determined aerodynamic model for model-based navigation on a delta-wing UAV

Abstract

Model-based navigation is a promising approach for autonomous navigation of small drones in challenging conditions such as GNSS denied flight scenarios. However, the lack of analysis of aerodynamic model structure for model-based navigation applications on delta-wing UAVs, characterized by a reduced number of control surfaces, has hindered its practical implementation. In this study, we propose a methodology for generalizing an aerodynamic model experimentally determined for a specific platform to a family of platforms sharing comparable physical characteristics by employing in-flight tuning. The experimental results show that the proposed methodology significantly improves navigation performance under GNSS outage, compared to traditional autonomous navigation approaches, for the model adapted to a second delta-wing platform. This indicates that the proposed methodology can be used to adapt aerodynamic models to different delta-wing UAV platforms of similar size, enabling reliable model-based navigation in challenging environments. This work contributes to the advancement of autonomous navigation technology for small drones, particularly in applications where GNSS signals are unavailable or unreliable.