21497 research outputs found
Sort by
Roger B Crosskey - Biography
Get PDFBiography of RAF cadet LAC Roger B Crosskey who was killed in a night flying accident on January 20, 1942, while training to be a pilot at 5BFTS. He is buried in the Commonwealth War Graves Commission British Plot. at Oak Ridge Cemetery, Arcadia
Orbital Maneuvers and Interplanetary Trajectory Design via Reinforcement Learning
No full textThis dissertation investigates the application of reinforcement learning (RL) to the design and optimization of low-thrust spacecraft trajectories, with an emphasis on autonomy, adaptability, and robustness in the presence of system uncertainties and unmodeled perturbations. Classical approaches to low-thrust trajectory design are predominantly grounded in optimal control theory, which relies on the availability of precise dynamical models and often requires problem-specific reformulation and solver tuning. While optimal control methods offer high accuracy under deterministic conditions, their sensitivity to stochastic disturbances and computational limitations in highly nonlinear or uncertain environments pose significant challenges for future autonomous space missions.
To address these challenges, this research proposes a reinforcement learning framework in which spacecraft learn to generate continuous low-thrust control actions through direct interaction with the environment. The spacecraft dynamics are formulated using the two-body problem equations of motion and Gauss’ variational equations expressed in modified equinoctial elements, allowing for efficient handling of low-thrust propulsion and long-duration transfers. Several RL algorithms, including Proximal Policy Optimization and Soft Actor-Critic, are implemented and evaluated across a diverse set of trajectory design problems: orbit-raising maneuvers, inclination change maneuvers, combined orbital maneuvers, and an asteroid rendezvous mission targeting near-Earth asteroid Apophis.
The RL agents are trained under both deterministic and stochastic conditions, with stochastic perturbations modeled as zero-mean Gaussian white noise accelerations to simulate realistic environmental disturbances such as solar radiation pressure, navigation errors, and control noise. The performance of the RL-generated solutions is rigorously assessed through direct comparison with classical optimal control results obtained using pseudospectral methods, as well as through extensive Monte Carlo simulations that quantify robustness and terminal accuracy under uncertainty. Across all case studies, the RL policies successfully generate feasible low-thrust trajectories that achieve the target orbital conditions, demonstrating resilience to perturbations and generalization to new initial conditions.
The RL-based solutions offer a distinct advantage in terms of flexibility and adaptability without requiring explicit knowledge of the perturbation models. These results suggest that reinforcement learning can serve as an effective complementary tool for trajectory design, particularly in scenarios where ground communication is limited or system uncertainties preclude the use of fully deterministic guidance strategies. Potential applications include autonomous fault recovery, initial transfer design, and support for onboard decision-making during deep space operations.
This dissertation contributes to the growing body of research on machine learning for space applications by systematically evaluating the strengths and limitations of RL-based guidance in low-thrust trajectory optimization, highlighting the feasibility of deploying such methods in future autonomous missions. The findings provide insight into reward function design, algorithm selection, and robustness assessment, laying the groundwork for continued exploration of learning-based methods in astrodynamics
Boeing Week Prep Session
No full textConnect with Boeing Representatives to prepare for Boeing Week with resume reviews and career advice
Characterization of Search Spaces and Effects on Machine Learning
No full textThe present status of the field of Machine Learning (ML) focuses on optimization of popular models. Rarely are the effects of the problem characteristics upon the solution algorithm studied. There exists no standard for knowing when to apply ML algorithms to a given problem or how to estimate the effectiveness of results. Focusing on the search space of problems, a rigorous study was conducted to generate an in-depth understanding of the impact of search space characteristics to the performance of a ML algorithm, specifically a Genetic Algorithm (GA). The effects of specific problem characteristics, represented via solution space characteristics, on the efficiency and solution quality of a GA were measured. The results allow researchers to adapt ML techniques and estimate quality of results provided the solution space characteristics can be determined and mapped to those in this study. As part of this study standardized terms for solution space characteristics were developed. Standardization of these terms, used without formal definition in the literature, allow for a uniform characterization of search spaces, which in turn allows the results of this study to be transferred to other domains
The unaltered pulsar: GRO J1750-27, a supercritical X-ray neutron star that does not blink an eye
Get PDFWhen accreting X-ray pulsars (XRPs) undergo bright X-ray outbursts, their luminosity-dependent spectral and timing features can be analyzed in detail. The XRP GRO J1750-27 recently underwent one such episode, during which it was observed with NuSTAR and monitored with NICER. Such a data set is rarely available, as it samples the outburst over more than 1 month at a luminosity that is always exceeding ∼5 × 1037 erg s−1 . This value is larger than the typical critical luminosity value, where a radiative shock is formed above the surface of the neutron star. Our data analysis of the joint spectra returns a highly (NH ∼ (5−8) × 1022 cm−2 ) absorbed spectrum showing a Kα iron line, a soft blackbody component likely originating from the inner edge of the accretion disk, and confirms the discovery of one of the deepest cyclotron lines ever observed, at a centroid energy of ∼44 keV corresponding to a magnetic field strength of 4.7×1012 G. This value is independently supported by the best-fit physical model for spectral formation in accreting XRPs which, in agreement with recent findings, favors a distance of 14 kpc and also reflects a bulk-Comptonization-dominated accretion flow. Contrary to theoretical expectations and observational evidence from other similar sources, the pulse profiles as observed by NICER remain remarkably steady through the outburst rise, peak and decay. The NICER spectrum, including the iron Kα line best-fit parameters, also remain almost unchanged at all probed outburst stages, similar to the pulsed fraction behavior. We argue that all these phenomena are linked and interpret them as resulting from a saturation effect of the emission from the accretion column, which occurs in the high-luminosity regime
The giant outburst of EXO 2030+375
Get PDFThe Be X-ray binary EXO 2030+375 went through its third recorded giant outburst from June 2021 to early 2022. We present the results of both spectral and timing analysis based on NICER monitoring, covering the 2−10 keV flux range from 20 to 310 mCrab. Dense monitoring with observations carried out about every second day and a total exposure time of ∼160 ks allowed us to closely track the source evolution over the outburst. Changes in the spectral shape and pulse profiles showed a stable luminosity dependence during the rise and decline. The same type of dependence has been seen in past outbursts. The pulse profile is characterized by several distinct peaks and dips. The profiles show a clear dependence on luminosity with a stark transition at a luminosity of ∼2×1036 erg s−1 , indicating a change in the emission pattern. Using relativistic raytracing, we demonstrate how anisotropic beaming of emission from an accretion channel with a constant geometrical configuration can give rise to the observed pulse profiles over a range of luminosities
Machine Learning for Near-Proximity Operations
Get PDFThe space economy is projected to grow to 1.8 trillion USD by 2030, according to the World Economic Forum and McKinsey & Company. As space operations become more common and diversified, the hazardous nature of the environment is worsened by the growing number of complex actors. Currently, ground segments manage these hazards and risks, but they are faced with potential operation challenges such as communications delays, communication issues due to single event upsets, and orbital perturbations. Additional environmental challenges that further complicate these operations might be satellite collisions, anti-satellite weaponry, orbital debris, and uncontrolled space. Furthermore, the increasing number of actors in space is likely to outpace the growth of personnel in this field and the infrastructure to support them. To address these challenges, one possible solution is the use of machine learning (ML) solutions in space to reduce the personnel needed and infrastructure to support them and increase autonomy. A key innovation in this field would be dynamic automated spacecraft coupling. In these scenarios, automated models would determine the optimal approach, implement the necessary actions to couple with the target object, and be capable of aborting if anomalies compromise the safety of the coupling. This study seeks to determine the viability of such a model using Proximal Policy Optimization along R-bar, V-bar, and Z-bar approaches in a 6-degree-of-freedom environment, implementing abort scenarios based on a predefined keep-out zone around the target objective
Miami Beach Postcard from early 1940s
No full textOne of a series of postcards bought by A W (Tony) Linfield, 5BFTS cadet on Course 18 (December 1943 - June 1944) while he was in Florida. Tony was a founder member of the 5BFTS Association and became the first Honorary Secretary. He donated his postcard collection to the 5BFTS Archives.https://commons.erau.edu/bfts-images-florida-ww2/1013/thumbnail.jp