Your tracks are too sparse and too noisy to trust an orbit. Short arcs, gaps of days between passes, and faint objects near the detection limit produce orbit solutions whose covariance does not reflect reality. I build estimation and image-processing pipelines that extract usable astrometry from marginal data and report honest uncertainty alongside the state. → Optical tracking and observatory automation

A satellite moved and you found out late. Distinguishing a genuine manoeuvre from mismodelled dynamics or a bad track is the difference between a real custody break and a false alarm. I develop manoeuvre-detection and pattern-of-life methods for geostationary and geosynchronous satellites, built on labelled optical data rather than assumptions. → Satellite manoeuvre detection

Your mission goes where GNSS does not reach. Beyond geostationary orbit and around the Moon, Earth-based positioning degrades or disappears, and ground tracking is a scarce, expensive resource. I design navigation architectures that hold accuracy autonomously — inter-satellite links, constellation geometry, and time synchronisation treated as one problem. → Lunar and cislunar navigation

You cannot say when it comes down, or what it belongs to. Reentry timing and conjunction screening both degrade when atmospheric density is modelled badly, and debris fragments arrive without a parent. I work on reduced-order density models driven by real ephemerides, and on classifiers that reconnect fragments to their breakup event. → Reentry and density prediction

Funding

• ARC Discovery Project DP260100821 — Lunar navigation. With Professor Andrew Dempster (UNSW ACSER, lead chief investigator) and Professor Junichiro Kawaguchi. Develops single-satellite Doppler positioning, autonomous on-board positioning of lunar navigation satellites, and time synchronisation for lunar PNT, with outcomes targeted at Australia’s robotic lunar mission.
• NSW Space Research Network Pilot Research Project (2025–2026). Pattern of Life Identification for Geosynchronous Satellites Using Transformer-Based AI Foundation Models, in collaboration with Macquarie University and the University of Adelaide.
• NSW Space Research Network Student Project (2025–2026). Automated Satellite Tracking Control at UNSW Observatory.

External responsibility

• Associate Editor, IEEE Transactions on Aerospace and Electronic Systems (Space Systems), from May 2026.
• IEEE Senior Member since 2023; panelist on the IEEE Region 10 Senior Member Panel, 2026.
• Organising committee, PNT 2026 (Australian Institute of Navigation) — coordinated the student presentation competition and facilitated the Lunar Navigation session.
• Contributing author on two NSW Space Research Network white papers, on Positioning, Navigation and Timing and on Space Domain Awareness.
• Academic Lead, AUStronauts — a UNSW Engineering Vertically Integrated Project developing a 3U CubeSat with an electrodynamic tether.

Results on the record

• Debris family classification improved to ROC-AUC 0.858, against 0.789 for the modified-equinoctial baseline, by combining three proper-element sets. Restoring orbital size information to the quaternion set lifted its accuracy from 0.31 to 0.60. arXiv:2512.08495
• Autonomous lunar orbit determination to approximately 11 m in 3-D position and 1.58 mm/s in velocity using inter-satellite links alone, with no direct ground support to the navigation satellite (CR3BP case); approximately 1 km and 10 mm/s under a high-fidelity dynamical model after a two-day ground-tracking handover. MPhil thesis, UNSW Sydney
• Lunar navigation constellation optimised for south-pole coverage, integrating a semi-analytical station-keeping model with Earth-GPS for time synchronisation and ephemeris generation. Acta Astronautica 239, 278–293 (2026)
• UNSW Observatory automated for remote satellite tracking, converting a heritage instrument into a taskable sensor for space object tracking and characterisation. Project page · thesis slides

Tracking campaign. You nominate the objects. The UNSW Observatory is tasked against them, and you receive astrometry, orbit solutions and a residual and accuracy report.

Manoeuvre and pattern-of-life assessment. You supply a GEO or LEO time series, or we source one. You receive a labelled event list and detection performance benchmarked against your current baseline.

Cislunar and lunar PNT trade study. You supply mission geometry and constraints. You receive a coverage and accuracy budget, and a navigation architecture recommendation.

Here are some photos with my colleagues and students.

Previous Next