Evidence-Based Robust Design of Deflection Actions for Near Earth Objects

This paper presents a novel approach to the robust design of deflection actions for Near Earth Objects (NEO). In particular, the case of deflection by means of Solar-pumped Laser ablation is studied here in detail. The basic idea behind Laser ablation is that of inducing a sublimation of the NEO surface, which produces a low thrust thereby slowly deviating the asteroid from its initial Earth threatening trajectory. This work investigates the integrated design of the Space-based Laser system and the deflection action generated by laser ablation under uncertainty. The integrated design is formulated as a multi-objective optimisation problem in which the deviation is maximised and the total system mass is minimised. Both the model for the estimation of the thrust produced by surface laser ablation and the spacecraft system model are assumed to be affected by epistemic uncertainties (partial or complete lack of knowledge). Evidence Theory is used to quantify these uncertainties and introduce them in the optimisation process. The propagation of the trajectory of the NEO under the laser-ablation action is performed with a novel approach based on an approximated analytical solution of Gauss’ Variational Equations. An example of design of the deflection of asteroid Apophis with a swarm of spacecraft is presented.

💡 Research Summary

The paper proposes a robust, evidence‑based design methodology for deflecting Near‑Earth Objects (NEOs) using solar‑pumped laser ablation. The concept relies on concentrating solar energy onto a laser system mounted on a formation of small spacecraft; the laser beam irradiates the asteroid surface, causing sublimation of material and generating a low, continuous thrust that gradually changes the asteroid’s trajectory.
To evaluate the performance of such a system, the authors develop an integrated model that couples (i) the low‑thrust dynamics of the asteroid under laser‑induced acceleration, (ii) a detailed spacecraft mass and power budget, and (iii) epistemic uncertainties affecting both the physical ablation process and the spacecraft technology.
The dynamical model uses Gauss’ variational equations expressed in non‑singular equinoctial elements. Because thousands of trajectory simulations are required for optimization, a novel analytical approximation called Finite Perturbative Elements in Time (FPET) is introduced. FPET treats the thrust vector as constant over short arcs of true longitude ΔL, expands the orbital element rates to first order in the thrust magnitude ε, and integrates analytically. Arc lengths are automatically adjusted according to the instantaneous thrust magnitude using a logarithmic scaling law, yielding a computational speed‑up of an order of magnitude (0.2–2 s per propagation versus ~30 s for full numerical integration) while preserving accuracy in the impact‑parameter b.
The spacecraft subsystem model accounts for a primary solar‑concentrating mirror, a secondary mirror, solar arrays, the laser, radiators, and harnesses. Masses are expressed as linear functions of areas, powers, and specific mass coefficients, with conservative margins (20 % for dry mass, 15 % for solar arrays, 25 % for mirrors, 50 % for lasers). Propellant mass is added as a fixed 10 % of the dry mass.
Uncertainties in parameters such as laser efficiency, solar‑array efficiency, material sublimation enthalpy, and reflectivity are represented using Evidence Theory. Each uncertain parameter is assigned a set of possible intervals (basic probability assignments) and associated belief and plausibility measures. This framework enables the propagation of epistemic uncertainty through the multi‑objective optimization, which simultaneously maximizes the impact‑parameter deviation b and minimizes total system mass. The optimization is performed with a modified NSGA‑II algorithm that evaluates belief‑based bounds on the objectives, producing a robust Pareto front. Sensitivity analysis identifies laser and solar‑array efficiencies as the dominant contributors to performance variability, while material properties have secondary effects.
A case study on asteroid 99942 Apophis (modified orbit to intersect Earth in 2036) demonstrates the methodology. The optimal design consists of a swarm of three to five identical spacecraft, each carrying a 1.2 m² primary mirror, a 0.8 m² secondary mirror, a 0.5 kW laser, and accompanying power and thermal subsystems. The total formation mass is about 1.8 tonnes. Simulations show that this configuration can increase the b‑parameter at the predicted impact epoch by more than 0.05 AU, effectively eliminating the collision risk.
Overall, the paper delivers a comprehensive, computationally efficient, and uncertainty‑aware framework for designing laser‑ablation NEO deflection missions, bridging the gap between high‑fidelity trajectory analysis and practical spacecraft engineering under epistemic ignorance.