Tight Integration of GNSS/LEO/INS in Dense Urban Environments

Yoji Takayama, Akihiro Osugi, Thyagaraja Marathe, Jérôme Leclère, Bryan Chan

Peer Reviewed

Abstract:	The surge of Low Earth Orbit (LEO) satellites for non-terrestrial networks is unprecedented, presenting a significant opportunity to carry out Positioning, Navigation, and Timing (PNT) from Low Earth Orbit, known as LEO PNT. LEO constellations for dedicated PNT services are planned to be deployed by companies and space agencies such as Xona Space Systems and the European Space Agency (ESA). This paper aims to address a significant gap in understanding how accuracy can be achieved in dense urban environments, where Global Navigation Satellite Systems (GNSS) and Regional Navigation Satellite Systems (RNSS) can be inaccurate, through the integration of GNSS, LEO, and Inertial Navigation System (INS) in a tightly coupled manner. Systems operating outdoors, such as robots, automobiles, and drones, are likely to use these methods concurrently for their PNT needs as they move towards automation. In this paper, we first introduce the visibility of LEO satellites. Their line of sight can be more obstructed than GNSS/RNSS satellites in dense urban environments, resulting in a few visible LEO satellites. Then, we describe the integration algorithm using GNSS, LEO, and INS in a tightly coupled manner. The integration of GNSS and INS has been well-studied so far. We further incorporate LEO into the integration of GNSS and INS based on the similarity of the system architectures between GNSS and LEO. The last part of this paper is devoted to examining the accuracy in dense urban environments when using the developed integration algorithm with a mixed virtual-real simulation program, which uses actual vehicle motion and simulated measurements based on the simulated sky view created by Google Earth™. The results show that the horizontal accuracy of the developed integration algorithm can be approximately half a meter in dense urban environments where the total number of directly visible satellites is about 6 on average including one LEO satellite. The key contribution of this study is to demonstrate that the tight integration of GNSS, LEO, and INS provides a promising solution for positioning in dense urban environments. This may suggest that the fast convergence of estimation can be achieved due to the faster motion of LEO satellites, even when, on average, only one LEO satellite is directly visible.
Published in:	Proceedings of the 38th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2025) September 8 - 12, 2025 Hilton Baltimore Inner Harbor Baltimore, Maryland
Pages:	2346 - 2359
Cite this article:	Takayama, Yoji, Osugi, Akihiro, Marathe, Thyagaraja, Leclère, Jérôme, Chan, Bryan, "Tight Integration of GNSS/LEO/INS in Dense Urban Environments," Proceedings of the 38th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2025), Baltimore, Maryland, September 2025, pp. 2346-2359. https://doi.org/10.33012/2025.20326
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In