We are witnessing a space renaissance. Tens of thousands of broadband low Earth orbit (LEO) satellites are expected to be launched by the end of this decade. These planned megaconstellations of LEO satellites along with existing constellations will shower the Earth with a plethora of signals of opportunity, diverse in frequency and direction. These signals could be exploited for navigation in the inevitable event that global navigation satellite system (GNSS) signals become unavailable (e.g., in deep urban canyons, under dense foliage, during unintentional interference, and intentional jamming) or untrustworthy (e.g., under malicious spoofing attacks). Nevertheless, to exploit “uncooperative” LEO satellites opportunistically for navigation, a number of incumbent challenges must be addressed, namely, the unknown nature of their signals, ephemerides, and clock errors. Recent advances in cognitive receiver design have shown the ability to extract navigation observables (pseudorange, Doppler, and/or carrier phase) from unknown LEO signals. This tutorial focuses on addressing the latter challenges via a framework termed STAN: simultaneous tracking and navigation. STAN estimates the navigating vehicle’s states simultaneously with the states of orbiting LEO satellites. STAN employs a cognitive receiver that exploits LEO satellite downlink signals to produce navigation observables, which are fused through an extended Kalman filter (EKF) to aid the vehicle’s inertial navigation system (INS) in a tightly coupled fashion. First, this tutorial presents the models governing the vehicle’s INS kinematics, LEO satellite dynamics, clock error dynamics, and LEO measurements (pseudorange, Doppler, and/or carrier phase). Next, the tutorial formulates the EKF’s state vector and details the EKF’s time and measurement updates. To demonstrate the efficacy of STAN, the tutorial presents simulation results showcasing an aerial vehicle navigating with unknown LEO satellites. The aerial vehicle is assumed to be equipped with an altimeter and a tactical-grade inertial measurement unit (IMU), navigating for 15.43 km in 300 s, in which GNSS signals were only available for the first 60 s. It is demonstrated that the final three-dimensional (3-D) position error and position root mean squared error (RMSE) of a typical tightly coupled GNSS-aided INS grows to 1,536 and 897 m, respectively. In contrast, the STAN framework with 77 LEO satellites (resembling Orbcomm, Iridium NEXT, and Starlink constellations) achieved a final 3-D position error and 3-D position RMSE of 15.2 and 7.3 m, respectively, with pseudorange measurements, and 37.1 and 10.6 m, respectively, with Doppler measurements. To demonstrate the efficacy of STAN in the real world, the tutorial presents experimental results of a ground vehicle navigating with multiconstellation LEO satellites. The vehicle traversed 4.15 km in 150 s, in which GNSS signals were only available for the first 80 s. It is shown that the final 3-D position error and 3-D position RMSE of the vehicle’s GNSS-aided INS with an industrial-grade IMU and an altimeter grew to 472.7 and 118.5 m, respectively. In contrast, the final 3-D position error and position RMSE of the STAN framework with signals from 2 Orbcomm, 1 Iridium NEXT, and 3 Starlink LEO satellites were 27.1 and 18.4 m, respectively.
Ad Astra: Simultaneous Tracking and Navigation With Megaconstellation LEO Satellites
IEEE Aerospace and Electronic Systems Magazine ; 39 , 9 ; 46-71
2024-09-01
2392276 byte
Article (Journal)
Electronic Resource
English
PERFORMANCE ANALYSIS OF SIMULTANEOUS TRACKING AND NAVIGATION WITH LEO SATELLITES
British Library Conference Proceedings | 2020
|British Library Conference Proceedings | 2021
|