Elisavetha (Lizzie) Sergeev

At the Geostationary Earth Orbit (GEO), routine observations lack the resolution to reveal physical characteristics, especially for smaller, discarded objects. To address this, we introduce a new classification system that succinctly characterises GEO objects by their control, regional residence and coplanarity to detect drifters. Applied over the entire catalogue from 1963 to 2025, the data shows that 36.3% of uncontrolled objects reside within the protected region containing operational satellites, with each 1◦ slot hosting 1.2 drifters on average, underscoring the value of routinely tracking uncontrolled traffic at GEO. To estimate the otherwise unobservable physical parameters (illuminated area and mass) of these drifting objects, we present an improved Two-Line Element (TLE) catalogue. To infer area and mass, a two-stage, coarse-to-fine parameter search is coupled to a High Precision Orbital Propagator (HPOP) and robust z-score fit metrics scaled by IQR. Matching the history of the eccentricity ground truth to the propagated result at various area-mass pairs yields an informed approximation. To validate an eccentricity-only approach, we show eccentricity fit errors dominate the dispersion, with the median within-object IQR ratios of 80.4 versus semi-major axis and 6.4 versus position-magnitude errors. The area and mass pair selected by minimising the eccentricity error was Pareto-optimal when compared against both position and semi-major axis errors and confirms our eccentricity-driven strategy is sufficient to infer illuminated area and mass. We also observe time-varying area-to-mass ratio, consistent with solar radiation pressure aliasing of rotational dynamics reported for high area-to-mass ratio debris. Drifting satellites with maximum observation cadence >5 days have on average 24.1% higher mean absolute position error, and beyond this threshold the error grows by 11.6% per additional unobserved day. Duplicate measurements at the initial epoch inflate the error by 23.9% for drifters on average. These statistics quantify how data hygiene and observation frequency directly influence poorer area/mass estimates by polluting our reference ground truth. The results indicate that regularised cadence, informed datasets with physical characteristics and clean initial conditions for GEO Protected Region (GEOPR) drifters can reduce position error by 92.2%. In conclusion, our update to the TLE catalogue with parameters and filters will enhance trajectory propagation and therefore risk forecasting at GEO.

Public Two-Line Element (TLE) catalogues enable repeatable studies of the Geostationary Earth Orbit (GEO) debris environment; however, their use in slot-level collision risk assessment is limited by simplified control volumes and the difficulty of linking congestion metrics directly to operators. Building on the longitude-resolved torus Cell Passage Event (CPE) framework, we present a TLE-driven pipeline that calibrates effective solar-radiation-pressure area-to-mass histories, propagates drifting objects with a high-fidelity numerical model, and replaces the idealised torus with two complementary, operator-relevant methods: (i) a hollow spherical shell aligned with formal GEO Protected Region (GEOPR) bounds and (ii) covariance-normalised proximity screening between drifting objects and maintained-slot references. For the controlled comparison, we analyse a common 92-day propagation window using 10-minute torus sampling and 5-minute hollow-shell and covariance sampling, with local Lagrange interpolation for event refinement. In a separate paired causal validation across 214 objects, joint area–mass inference reduced the median next-interval three-dimensional position error from 4.54 km to 4.31 km, with improvement for 70.1% of objects. The longitude-resolved baseline uses 1.0◦ cells and a 50.0 km torus minor radius, yielding 4.21 × 104 CPEs across 312 drifters, with a median of 20 occupied longitude cells per detected drifter. Replacing the torus with the hollow-shell GEOPR volume records 2.62×105 passages across 720 drifting objects, with a median of 90 occupied cells per drifter. The hollow shell therefore records 6.23 times as many events and 130.8% more contributing objects. Despite this broader coverage, the longitude distributions retain a Spearman rank correlation of ρ = 0.788 and identify the principal concentrations associated with the geopotential gravitational wells near 75◦E and 105◦W. Across sensitivity windows of three to seven months, the hollow-to-torus event ratio remains between 6.16 and 6.23, while longitude rank agreement ranges from ρ= 0.788 to 0.839, indicating that the principal spatial structure is persistent. The second method replaces the previous fixed-radius proximity sphere with an uncertainty-normalised, operator-centred screen. Empirical diagonal Radial, Transverse and Normal (RTN) covariance proxies are estimated from held-out propagated TLE residuals for each drifter and from historical scatter about each maintained-slot reference. Samples satisfying Mahalanobis separation D ≤ 3 are retained, and consecutive retained samples are merged into a single encounter. The 92-day analysis retains 1.23 × 105 drifter–slot encounters involving 918 drifters, 442 maintained objects and 25,493 unique pairs across 234 occupied 1.0◦ longitude cells. Of these encounters, 3,645 (2.97%) have physical miss distances below 100 km and 82 fall below 10 km. The encounter burden is concentrated near approximately 85◦E and 140◦W, translating uncontrolled drift into an operator-centred prioritisation signal for higher-fidelity follow-up. Across the operator-relevant methods, drifter contribution becomes more concentrated as the screening geometry broadens: the top 10% of drifters account for 14.3% of torus events, 28.2% of hollow-shell passages and 29.4% of covariance-normalised encounters. Alternative spatial, velocity and consequence-weighting kernels preserve the principal hollow-shell longitude ranking, with Spearman correlations of ρ= 0.982–0.985 relative to the hollow geometry control. This indicates that protected-region occupancy governs the global congestion pattern, while alternative risk kernels primarily refine local priorities. These results support a robust, reproducible GEO screening framework that identifies persistent longitudinal concentrations and a manageable watchlist of high-contribution drifters, motivating targeted monitoring, higher-cadence updates and uncertainty-aware reruns.

Biomimetic and Bioinspired design is not only a potent resource for roboticists looking to develop robust engineering systems or understand the natural world. It is also a uniquely accessible entry point into science and technology. Every person on Earth constantly interacts with nature, and most people have an intuitive sense of animal and plant behavior, even without realizing it. The Natural Robotics Contest is novel piece of science communication that takes advantage of this intuition, and creates an opportunity for anyone with an interest in nature or robotics to submit their idea and have it turned into a real engineering system. In this paper we will discuss the competition's submissions, which show how the public thinks of nature as well as the problems people see as most pressing for engineers to solve. We will then show our design process from the winning submitted concept sketch through to functioning robot, to offer a case study in biomimetic robot design. The winning design is a robotic fish which uses gill structures to filter out microplastics. This was fabricated into an open source robot with a novel 3D printed gill design. By presenting the competition and the winning entry we hope to foster further interest in nature-inspired design, and increase the interplay between nature and engineering in the minds of readers.

Biomimetic and bioinspired design is not only a potent resource for roboticists looking to develop robust engineering systems or understand the natural world. It is also a uniquely accessible entry point into science and technology. Every person on Earth constantly interacts with nature, and most people have an intuitive sense of animal and plant behaviour, even without realizing it. The Natural Robotics Contest is novel piece of science communication that takes advantage of this intuition, and creates an opportunity for anyone with an interest in nature or robotics to submit their idea and have it turned into a real engineering system. In this paper we will discuss the competition's submissions, which show how the public thinks of nature as well as the problems people see as most pressing for engineers to solve. We will then show our design process from the winning submitted concept sketch through to functioning robot, to offer a case study in biomimetic robot design. The winning design is a robotic fish which uses gill structures to filter out microplastics. This was fabricated into an open source robot with a novel 3D printed gill design. By presenting the competition and the winning entry we hope to foster further interest in nature-inspired design, and increase the interplay between nature and engineering in the minds of readers.