Propulsionless planar phasing of multiple satellites using deep reinforcement learning

This work creates a framework for solving highly non-linear satellite formation control problems by using model-free policy optimisation deep reinforcement learning (DRL) methods. This work considers, believed to be for the first time, DRL methods, such as advantage actor-critic method (A2C) and proximal policy optimisation (PPO), to solve the example satellite formation problem of propellantless planar phasing of multiple satellites. Three degree-of-freedom simulations, including a novel surrogate propagation model, are used to train the deep reinforcement learning agents. During training, the agents actuated their motion through cross-sectional area changes which altered the environmental accelerations acting on them. The DRL framework designed in this work successfully coordinated three spacecraft to achieve a propellantless planar phasing manoeuvre. This work has created a DRL framework that can be used to solve complex satellite formation flying problems, such as planar phasing of multiple satellites and in doing so provides key insights into achieving optimal and robust formation control using reinforcement learning.     

Data Channel Monitoring by Correlation Techniques

The technique of analyzing a set of channel parameters is discussed. Pseudorandom probing (for low-level noncorrelative interference with a typical user) and cross-correlation analysis are employed in the system described. The transfer function of this communications channel is convolved with the autocorrelation function of the probe signal; hence, analysis of the cross-correlation function at the receiver yields channel phase and amplitude parameters on a discrete basis. Implementation of these concepts with a demonstrable engineering model has been accomplished, and laboratory test results are presented.     

The Swift X-Ray Telescope

he Swift Gamma-Ray Explorer is designed to make prompt multiwavelength observations of gamma-ray bursts (GRBs) and GRB afterglows. The X-ray telescope (XRT) enables Swift to determine GRB positions with a few arcseconds accuracy within 100 s of the burst onset. The XRT utilizes a mirror set built for JET-X and an XMM-Newton/EPIC MOS CCD detector to provide a sensitive broad-band (0.2–10 keV) X-ray imager with effective area of > 120 cm² at 1.5 keV, field of view of 23.6 × 23.6 arcminutes, and angular resolution of 18 arcseconds (HPD). The detection sensitivity is 2×10⁻¹⁴ erg cm⁻² s⁻¹ in 10⁴ s. The instrument is designed to provide automated source detection and position reporting within 5 s of target acquisition. It can also measure the redshifts of GRBs with Fe line emission or other spectral features. The XRT operates in an auto-exposure mode, adjusting the CCD readout mode automatically to optimize the science return for each frame as the source intensity fades. The XRT will measure spectra and lightcurves of the GRB afterglow beginning about a minute after the burst and will follow each burst for days or weeks. Dedicated to David J. Watson, in memory of his valuable contributions to this instrument.