A Anoop, BosonQ Psi Corp; Kandula Eswara Sai Kumar, BosonQ Psi Corp; Abhishek Chopra, BosonQ Psi Corp; Rut Lineswala, BosonQ Psi Corp
https://doi.org/10.64861/IEFW2406
Keywords: Satellite Conjunction Analysis, Collision Avoidance, Space Situational Awareness (SSA), Space Domain Awareness (SDA), Quantum‑Inspired Optimization, Conjunction Events, Space Traffic Management
Abstract:
The rapid expansion of satellite deployments in Earth orbit, particularly within the Low Earth Orbit (LEO) regime, has significantly increased the operational complexity of space traffic management and heightened the risk of accidental satellite collisions. Modern satellite constellations, including large-scale commercial deployments, have contributed to an increasingly congested orbital environment populated not only by active spacecraft but also by defunct satellites, spent rocket bodies, and orbital debris. As a consequence, the frequency of close approaches between space objects, commonly referred to as conjunction events, has risen sharply, making accurate and computationally efficient conjunction analysis a critical requirement for collision avoidance, mission safety, and long-term orbital sustainability. In this context, timely identification of the point of closest approach (PCA), characterized by the time of closest approach (TCA) and distance of closest approach (DCA), forms the analytical foundation for informed decision-making regarding evasive maneuvers.
The primary goal of this research is to investigate and demonstrate the effectiveness of a Quantum-Inspired Evolutionary Optimization (QIEO) algorithm for solving satellite conjunction analysis problems. Specifically, the study aims to evaluate whether QIEO can accurately identify both global and local minima of the relative separation distance between orbiting satellites, while achieving substantially reduced computational convergence times when compared to conventional techniques. The research focuses on two key objectives: first, the precise determination of the global point of closest approach between a given pair of satellites within a specified analysis window; and second, the detection of all conjunction events occurring within a user-defined distance threshold, corresponding to local minima of the relative distance function that represent potentially hazardous close encounters.
Conjunction analysis is inherently challenging due to the highly nonlinear, oscillatory, and multimodal nature of satellite relative motion. The relative separation function between two satellites exhibits multiple local minima arising from the periodic characteristics of orbital dynamics. Traditional gradient-based optimization techniques are therefore unsuitable, as they are prone to becoming trapped in suboptimal local minima and require analytically complex gradient computations. Brute-force approaches, while conceptually straightforward, suffer from rapidly increasing computational cost as temporal resolution is refined. Operational methods such as the Alfano algorithm address this challenge through cubic interpolation over discrete time samples, providing accurate solutions but at the expense of comparatively high computational effort, particularly when extended to multi-event analysis.
To address these limitations, this study adopts a metaheuristic optimization framework centered on QIEO. The methodological approach begins with orbit propagation using the SGP4 model, which computes satellite position and velocity states as a function of time based on Two-Line Element (TLE) data. From these propagated states, a time-dependent relative separation function is constructed, representing the Euclidean distance between satellite position vectors at each time step. The optimization problem is then formulated such that the decision variable corresponds to time within the analysis window, and the objective function to be minimized is the relative separation distance.
The research methodology evaluates three distinct solution approaches: Quantum-Inspired Evolutionary Optimization, a classical Genetic Algorithm (GA), and the Alfano method. In the QIEO framework, candidate solutions are encoded using probabilistic representations analogous to quantum bits, where each gene represents the probability of a binary state rather than a fixed value. Evolution of the population is governed by quantum-inspired rotation gate operators, which adjust these probabilities to guide the search toward regions of improved fitness. This approach eliminates the need for crossover and mutation operators used in classical genetic algorithms and enables highly parallelizable optimization with fewer function evaluations.
The GA implementation serves as a baseline evolutionary method, employing binary encoding, selection, crossover, and mutation to explore the search space. The Alfano method is included as a widely accepted operational benchmark, relying on cubic interpolation of discrete separation samples to estimate the time and distance of closest approach with high accuracy. All three methods are applied to identical problem instances to ensure fair comparison in terms of solution accuracy and convergence behavior.
The methodology is applied to two representative satellite conjunction scenarios derived from publicly available orbital data. For each case, the algorithms are evaluated on their ability to identify the global minimum corresponding to the closest approach event, as well as their effectiveness in detecting all local minima within predefined distance thresholds. Performance metrics include convergence time, agreement with reference solutions, and consistency in identifying multiple conjunction events.
The expected outcomes of this research demonstrate that QIEO is capable of achieving accuracy comparable to established methods such as the Alfano algorithm, while significantly outperforming both Alfano and GA in terms of computational efficiency. The results indicate that QIEO consistently converges to solutions that match reference TCA and DCA values to significant digits, while requiring an order of magnitude less computation time in several tested scenarios. Furthermore, QIEO successfully identifies all relevant local minima corresponding to conjunction events within specified thresholds, highlighting its robustness in multimodal optimization landscapes.
By combining high solution fidelity with substantially reduced convergence times, the proposed QIEO-based approach offers a compelling alternative for real-time or near-real-time conjunction analysis applications. The findings suggest that QIEO is particularly well-suited for scaling to multi-satellite conjunction screening problems, where computational efficiency is paramount. Overall, this research underscores the potential of quantum-inspired metaheuristic algorithms as powerful tools for next-generation space situational awareness and collision avoidance systems in increasingly congested orbital environments.
Date of Conference: September 15-18, 2026
Track: Conjunction/RPO