GPS Denied Navigation Using Meta-Image Objects From Georeferenced Maps

B.A. Schnaufer, K. Celik, J. Nadke, P. Hwang, A. Somani

Abstract:	Accurate position, velocity and attitude (PVA) is crucial for the success of Unmanned Aerial Vehicle (UAV) military operations. GPS is a critical sensor for such platforms due to its accuracy, global coverage and small hardware footprint. However, GPS is subject to denial in UAV operational environments due to blockage or RF interference. When GPS is not available, PVA performance from other inertial and air data sensor equipage alone is not sufficient, especially for small UAVs. However, UAVs can be equipped with small, low-cost, good quality image sensors, which can be used for navigation as well as for surveillance. In recent years image navigation for UAVs has received significant attention and much progress has been made in developing algorithms that can be implemented in real-time and that provide acceptable accuracy performance. However, some issues still remain such as the robustness and reliability of an image-based solution, as well as storage requirements for reference maps that can be used for absolute navigation. To address these limitations of GPS Denied UAV image navigation we have developed a novel, automated algorithm called MINA (Meta Image Navigation Augmenters), which is a synergy of machine-vision and machine-learning algorithms for absolute navigation georeferenced map objects. As opposed to existing image map matching algorithms, MINA utilizes geo-referenced vector map data, such as found with open sources like OpenStreetMap, in conjunction with real-time optical imagery from an on-board, monocular camera to augment the UAV navigation computer. For vector maps, which are much more compact than image maps, a map for a very large operational region can be stored on-board the aircraft in an optimized XML database. The onboard map is assumed to be imperfect, incomplete, heterogeneous, and created by diverse set of independent volunteers. MINA interprets and renders spatial areas represented by the vector map data classifications which are, in turn, correlated against real-time aerial imagery to generate correspondences that provide position measurement information for enhanced PVA state determination. Image navigation robustness is enhanced by tracking and correlating more complex image objects in the real-time image stream, which are less susceptible for false matches and provide rotational as well positional measurement information. MINA has been experimentally validated with both actual flight data and flight simulation data and results are presented in the paper. The MINA algorithm is effective for augmenting UAV GPS Denied navigation performance over very large regions of operation in a data efficient manner.
Published in:	Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013) September 16 - 20, 2013 Nashville Convention Center, Nashville, Tennessee Nashville, TN
Pages:	746 - 755
Cite this article:	Schnaufer, B.A., Celik, K., Nadke, J., Hwang, P., Somani, A., "GPS Denied Navigation Using Meta-Image Objects From Georeferenced Maps," Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013), Nashville, TN, September 2013, pp. 746-755.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In