Wide-Area UAV Networks Cooperative Positioning Algorithm Based on Information Geometry

This letter discusses the improvement of cooperative positioning frameworks for wide-area unmanned aerial vehicle (UAV) networks under the ranging measurement alone. The cooperative nodes are far apart in the wide-area UAV networks, which makes the angle measurement methods result in significant 3-D positioning errors. Therefore, only the ranging information between nodes is utilized to optimize the positioning results. To improve the accuracy and computation speed of the cooperative positioning algorithm, the ranging measurement-based manifold gradient fusion (RM-MGF) method is proposed. An error propagation model for cooperative nodes based on dilution of precision (DOP) is derived, making the error estimation more accurate. Furthermore, the Riemannian manifold gradient corresponding to the measurement network is applied to the improvement of positioning accuracy. The experiment results verify that the proposed algorithm has the best positioning accuracy and lower computational complexity.