RISK CDT - Improving conjunction analysis using a combination of high-fidelity astrodynamics models and advanced numerical Bayesian methods

PLEASE APPLY ONLINE TO THE SCHOOL OF ENGINEERING, PROVIDING THE PROJECT TITLE, NAME OF THE PRIMARY SUPERVISOR AND SELECT THE PROGRAMME CODE "EGPR" (PHD - SCHOOL OF ENGINEERING)

This is a project within the multi-disciplinary EPSRC and ESRC Centre for Doctoral Training (CDT) on Quantification and Management of Risk & Uncertainty in Complex Systems & Environments, within the Institute for Risk and Uncertainty. The studentship is granted for 4 years and includes, in the first year, a Master in Decision Making under Risk & Uncertainty. The project includes extensive collaboration with prime industry to build an optimal basis for employability.

With both the number of operational satellites and the amount of orbital debris increasing, the potential for collisions between orbital objects is rising. Conjunction analysis is the calculation of the probability that two orbital objects collide. The calculation of such probabilities is necessarily a function of the uncertain knowledge of objects’ positions and velocities, as derived from observations from, for example, telescopes and long-range radars. The geometry concerned means that the uncertainty is typically articulated as a “banana-shaped” probability distribution. While it is possible to approximate such a distribution as a Gaussian (or as a Gaussian mixture), such an approximation results in errors in the probability calculated in the conjunction analysis: the real probability is mis-estimated. This mis-estimation means that the operators of satellites have to either ignore valid alerts or expend fuel (and so reduce mission lifetime) unnecessarily. This project will investigate the ability to capitalise on recent advances in both accurately modelling the forces experienced by orbital objects (as developed extensively by UCL) and devising efficient numerical Bayesian methods (as developed extensively by researchers at the University of Liverpool). The numerical Bayesian methods involved will include particle filters, but also more recent extensions (e.g., the Sequential Monte Carlo (SMC) sampler), which are better placed to handle the near-deterministic trajectories that will become increasingly relevant as more sophisticated and accurate models for the forces are considered. The focus for the project is on working with UCL and Dstl. However, there is scope for the student to also interact with a number of other organisations (e.g., researchers working at or for: the UK RAF, the US government and insurance industry) who are interested in this problem and with whom the team have existing relationships.