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.     

Electromagnetic waves observed by Venus Express at periapsis: Detection and analysis techniques

The magnetometer on Venus Express was designed to be able to obtain 128 Hz samples of the magnetic field from two sensors in a gradiometer configuration. This mode is used around periapsis to determine whether the signals reported at low altitudes near 100 Hz, had the properties of electromagnetic waves generated by electric discharges in the Venus atmosphere. The lack of a magnetic cleanliness program and the shortness of the magnetometer boom make this a challenging measurement. Fortunately the signals are sufficiently strong that they can be easily resolved with rather straightforward analysis techniques.     

Revised upper limit on the internal magnetic moment of Venus

Over four Venus years of low altitude nightside PVO magnetometer observations are used to establish a new upper limit for the magnetic moment of Venus. Improvements over previous studies include data coverage and new instrument calibration information. The upper limit on an internal dipole moment is determined to be 8.4 × 10¹⁰ T m³.     

Numerical prediction and wind tunnel experiment for a pitching unmanned combat air vehicle

The low-speed flowfield for a generic unmanned combat air vehicle (UCAV) is investigated both experimentally and numerically. A wind tunnel experiment was conducted with the Boeing 1301 UCAV at a variety of angles of attack up to 70 degrees, both statically and with various frequencies of pitch oscillation (0.5, 1.0, and 2.0 Hz). In addition, pitching was performed about three longitudinal locations on the configuration (the nose, 35% MAC, and the tail). Solutions to the unsteady, laminar, compressible Navier–Stokes equations were obtained on an unstructured mesh to match results from the static and dynamic experiments. The computational results are compared with experimental results for both static and pitching cases. Details about the flowfield, including vortex formation and interaction, are shown and discussed, including the non-linear aerodynamic characteristics of the vehicle.     

Temperature comparison between CHAMP radio occultation and TIMED/SABER measurements in the lower stratosphere

Global Positioning System (GPS) receiver on the CHAllenging Mini-satellite Payload (CHAMP) and the Sounding of the Atmosphere using Broadband Emission Radiometry (SABER) instrument, one of four on board the TIMED satellite, provide middle atmosphere temperature profiles by Radio Occultation (RO) and limb viewing infrared emission measurements, respectively. These temperature profiles retrieved by two different techniques in the stratosphere are compared with each other using more than 1300 correlative profiles in March, September and December 2005. The over-all mean differences averaged over 15 and 35 km are approximately −2 K and standard deviation is less than 3 K. Below 20 km of altitude, relatively small mean temperature differences ∼1 K are observed in wide latitudinal range except for June (during the SABER nighttime observation). In the middle to low latitudes, between 30°S and 30°N, the temperature difference increases with height from ∼0–1 K at 15 km, to ∼−4 K at 35 km of altitude. Large temperature differences about −4 to −6 K are observed between 60°S and 30°N and 31–35 km of altitude for all months and between 0° and 30°N below 16 km during June (nighttime).     

ISEE-1 and 2 observations of field-aligned currents in the distant midnight magnetosphere

Magnetic field measurements obtained in the nightside magnetosphere by the co-orbiting ISEE-1 and 2 spacecraft have been examined for signatures of field-aligned currents (FAC). Such currents are found on the boundary of the plasma sheet both when the plasma sheet is expanding and when it is thinning. Plasma sheet boundary layer current structure and substorm associated dynamics can be determined using the two spacecraft, although for slow traversals of the FAC sheet the spatial/temporal ambiguity is still an issue. We often find evidence for the existence of waves on the plasma sheet boundary, leading to multiple crossings of the FAC sheet. At times the boundary layer FAC sheet orientation is nearly parallel to the X-Z GSM plane, suggesting ‘protrusions’ of plasma sheet into the lobes. The boundary layer current polarity is, as expected, into the ionosphere in the midnight to dawn local time sector, and outward near dusk. Current sheet thicknesses and velocities are essentially independent of plasma sheet expansion or thinning, having typical values of 1500 km and 20–40 km/s respectively. Characteristic boundary layer current densities are about 10 nanoamps per square meter.     

Alfvénic Electron Acceleration in Aurora Occurs in Global Alfvén Resonosphere Region

Auroral emission caused by electron precipitation (Hardy et al., 1987, J. Geophys. Res. 92, 12275–12294) is powered by magnetospheric driving processes. It is not yet fully understood how the energy transfer mechanisms are responsible for the electron precipitation. It has been proposed (Hasegawa, 1976, J. Geophys. Res. 81, 5083–5090) that Alfvén waves coming from the magnetosphere play some role in powering the aurora (Wygant et al., 2000, J. Geophys. Res. 105, 18675–18692, Keiling et al., 2003, Science 299, 383–386). Alfvén-wave-induced electron acceleration is shown to be confined in a rather narrow radial distance range of 4–5 R _E (Earth radii) and its importance, relative to other electron acceleration mechanisms, depends strongly on the magnetic disturbance level so that it represents 10% of all electron precipitation power during quiet conditions and increased to 40% during disturbed conditions. Our observations suggest that an electron Landau resonance mechanism operating in the “Alfvén resonosphere” is responsible for the energy transfer.