Debris collision avoidance maneuver optimization (CAMO) for satellite constellations

The exponential growth of the orbital debris population in Near-Earth space poses a significant threat to the sustainability of current and future satellite constellations. Traditional collision avoidance strategies, which typically rely on single-impulse maneuvers executed in response to ground-based warnings, often suffer from high propellant costs and operational inefficiencies due to late detection and reaction times. This thesis proposes and validates an autonomous, multi-objective optimization framework for collision avoidance maneuvers (CAMs) tailored for Medium Earth Orbit (MEO) constellations, specifically the Global Positioning System (GPS). The core of this research is the development of a “Hybrid Three-Burn Maneuver” strategy that ensures a closed-loop trajectory, returning the satellite precisely to its nominal station-keeping slot after evading the threat. The optimization engine utilizes the Non-dominated Sorting Genetic Algorithm II (NSGA-II) to simultaneously minimize collision probability (Pc) and total velocity change (∆V ). A high-fidelity simulation environment was constructed in MATLAB, incorporating J2-perturbed dynamics for debris and Keplerian propagation for satellites to capture realistic relative motion and nodal drift. The Probability of Collisionis computed using a robust K-series expansion method, enabling computationally efficient and numerically stable risk assessment. The framework was tested against six high-risk conjunction scenarios identified within a simulated GPS constellation, including a critical head-on encounter with a 221-meter miss distance. A parametric study was conducted across three temporal regimes: Strategic (> 8 hours warning), Operational (∼ 3 hours), and Tactical (10 minutes). Key findings indicate: 1. The Cost of Delay: There is a severe nonlinear relationship between maneuver warning time and fuel consumption. Strategic maneuvers executed hours in advance require approximately 0.5 m/s of ∆V , whereas emergency tactical maneuvers require over 8.0 m/s—an 18-fold increase in fuel cost representing a power-law scaling with reduced warning time. 2. Quarter-Period Optimal Time Scale: When unconstrained, the optimization algorithm consistently converges to a maneuver duration of k ≈ T /4 (where T is the orbital period), representing the fundamental optimal time scale for closed-loop collision avoidance in circular orbits. For GPS satellites with T = 43,080 s, this yields k ≈ 10,800 s (3 hours). This convergence occurs because the three-burn return-to-station constraint can only be exactly satisfied when 2k = nT /2 (where n is a positive integer), with n = 1 providing the minimum-energy solution. Maneuver durations significantly below T /4 enter the hyperbolic scaling regime, while durations above T /4 yield no additional fuel savings. 3. Algorithm Robustness: The hybrid evolutionary algorithm successfully identified safe trajectories (Pc < 10−6) for all test cases, demonstrating its capability to handle diverse encounter geometries. 4. Operational Viability: The proposed autonomous system enables a “low-energy drift” avoidance mode that is functionally unavailable to reactive ground-based systems, potentially extending satellite operational lifetimes by preserving critical station-keeping propellant. This research provides a quantitative basis for the implementation of onboard autonomous conjunction assessment and maneuver planning, offering a pathway to significantly enhance the resilience and longevity of critical space infrastructure.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Aerospace Engineering

Grantor

University of Illinois Urbana-Champaign

Year dc:date

2025

Author dc:creator

• Chandramukhi, Poornadithya

Contributors dc:contributor

• Coverstone, Victoria L

• Collision Avoidance, Satellite Constellations, Orbital Debris, Trajectory Optimization, Space Situational Awareness, Autonomous Systems, Multi-Objective Optimization

Statement dc:rights

• Copyright 2025 Poornadithya Chandramukhi

Language dc:language

en

Handle dc:identifier

https://hdl.handle.net/2142/132699

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/132699