Improving LEO Satellite Onboard SPP Orbits with Dynamic Models

Abstract

Low Earth Orbit (LEO) satellites are used for various applications, such as geophysical research, global communication, global navigation, and precise orbit determination. Different methods are available for LEO satellite orbit determination. This study aims to enhance LEO satellite low-accuracy Single Point Positioning (SPP) orbits on the ground by utilizing dynamic models. The study evaluates this enhancement by introducing errors in different directions of the SPP orbits and utilizing various dynamic models (including stochastic velocity pulses) for smoothing. The findings indicate that applying appropriate stochastic pulses can well improve the SPP orbits. A spacing of 1.5 hours of the pulses has shown to be a good option for reducing the STD of the orbital errors. Additional Solar Radiation Pressure (SRP) parameters are suggested to improve the orbits further when the spacing of the pulses is longer than 1.5 hours. Among them, the constant term in the along-track direction is shown to be essential. In the noise-only case, the SPP orbits can be reduced from meter-level to dm-level by applying appropriate SRP parameters and 1.5 h stochastic pulses. For SPP orbits containing large offsets, limitations exist in the improvements. Offsets in the SPP orbital errors often result in an offset in the along-track direction that is difficult to be removed. The study suggests a useful method to improve the accuracy and bridge gaps of the onboard determined orbits, which are of low accuracy and can only be transferred back to the ground with the corresponding Cartesian coordinates due to limited resources.