MATEC Web of Conferences (Jan 2019)

COTS based GNSS Receiver with Precise Point Positioning for CubeSats

  • Pandele Alexandru,
  • Cherciu Costel,
  • Trușculescu Marius,
  • Drăgășanu Claudiu,
  • Mihai Sergiu-Ştefan

DOI
https://doi.org/10.1051/matecconf/201930407009
Journal volume & issue
Vol. 304
p. 07009

Abstract

Read online

The paper presents the work towards developing a COTS based GNSS receiver and integrating Precise Point Positioning algorithms to facilitate close proximity operations for CubeSats in formation flying and during docking or rendezvous manoeuvres. We initially present the driving requirements identified for these types of missions. Besides fitting a standard CubeSat, the receiver has to weigh less than 0.3 kg, consume less than 5 W and be multi-frequency and multi-onstellation. Next, follows the identification of a commercial off the shelf GNSS receiver that can be easily customized to fit the basic requirements of a GNSS space receiver and on which Precise Orbit Determination (POD) algorithms can be implemented. For the start of the activity three commercial receivers were selected as proposed candidates to be traded off regarding the degree to which they fulfil the requirements, the degree of openness and of manufacturer support. A COTS microcontroller shall be then selected to control the operation of the COTS receiver. We then expand on the proposed general architecture of the system from COTS modules to their integration philosophy, with a discussion on the means of delivering the correction factors to the receiver. PPP corrections are expected to be delivered either via ground stations or via the geostationary satellite based commercial services. The PPP algorithms are to be implemented on the microcontroller, which will also try to maximize the availability of a precise PVT solution by incorporating a neural network fed by an orbit propagator and the PPP algorithm. The neural network shall estimate a precise position whenever PPP corrections are not available. The training of the neural network shall be done on the ground, allowing for a small footprint on board. A preliminary design of the hardware and the planned qualification plan is concluding the work.