The Potential of LEO Satellite-Based Opportunistic Navigation for High Dynamic Applications

Abstract

Resilient navigation in Global Navigation Satellite System (GNSS)-degraded and -denied environments is becoming more and more required for many applications. It can typically be based on multi-sensor data fusion that relies on alternative technologies to GNSS. In this work, we studied the potential of a low earth orbit (LEO) satellite communication system for a high-dynamic application, when it is integrated with an inertial measurement unit (IMU) and magnetometers. We derived the influence of the main error sources that affect the LEO space vehicle (SV) Doppler-based navigation on both positioning and attitude estimations. This allowed us to determine the best, intermediate and worst cases of navigation performances. We show that while the positioning error is large due to large orbit errors or high SV clock drifts, it becomes competitive with that of an inertial navigation system (INS) based on a better quality IMU if precise satellite orbits are available. On the other hand, the attitude estimation tolerates large orbit errors and high SV clock drifts. The obtained results suggest that LEO SV signals, used as signals of opportunity for navigation, are an attractive alternative in GNSS-denied environments for high dynamic vehicles.

Practical Use of Starlink Downlink Tones for Positioning

“…For comparison, we ran the filter by removing the frequency error from the filter state vector; we obtained a position error that exceeds 10 km (3D). This result is compliant with the expectation carried out in [ 21 ] showing that Doppler-based positioning is very sensitive to orbit errors. Therefore, the frequency error state has captured most of the satellites’ orbit and clock errors, preventing their total propagation into the user position state.…”

“…The range rate is the opposite sign of the Doppler shift times of the wavelength, i.e., . We have shown in [ 21 ] that the k th-satellite range rate error at time , , which is the difference between the measured range rate and the predicted one, can be expressed as a function of the user’s position error , the user’s velocity error , the user clock bias error , and the user clock drift error : where is the range rate noise, and is the line-of-sight unit vector between the receiver and the satellite given by: …”

“…In (16), we have added the term , representing the k th-satellite clock drift range- rate, which we purposely not included in [ 21 ]. The tropospheric induced frequency error is assumed to have been compensated based on a model (for instance, the MOPS [ 32 ]).…”

“…It can be based on the carrier phase [ 18 ] or on Doppler-shift measurements [ 15 , 16 ]. A second class performs a coupling between LEO satellite measurements and an IMU (gyroscopes and accelerometers) [ 19 , 20 ] with magnetometers [ 21 ]. In general, tight coupling is considered due to the weak satellite visibility from operational LEO systems that makes it difficult to determine a standalone positioning.…”

Generating Low-Earth Orbit Satellite Attitude Maneuver Profiles Using Deep Neural Networks

“…Because the demand for high-agility satellites that maximize the observation time is increasing, many satellites that minimize the maneuver time between the target pointings and attitude error at the initiation of the mission are being developed [ 9 , 10 , 11 ]. Additionally, as demand for satellite imagery for urgent missions (e.g., disaster response) increases, the need to maximize mission performance through rapid maneuvers also increases [ 12 ].…”