arXiv subjects

Euihyeon Choi

Publications and source records attributed to Euihyeon Choi.

3 recordsLinked to original sources

arXiv2026

Integrated Routing and Trajectory Design for Satellite Servicing

This paper studies the integrated spacecraft routing and trajectory optimization problem for satellite servicing missions involving partial en-route propellant replenishment. Unlike terrestrial routing problems, spacecraft operate in a dynamic environment, and we need to optimize the spacecraft routing over a network with nonlinear and time-dependent trajectory costs. In this paper, we tackle this problem using two different formulations. The first formulation, referred to as the arc-based formulation, defines variables based on arcs and employs an iterative decoupling scheme that alternates between mixed-integer linear programming and sequential nonlinear trajectory optimization. The second formulation, referred to as the path-based formulation, defines variables based on paths/routes and leverages column generation and a labeling algorithm to accelerate the identification of promising routes. Through a geosynchronous satellite servicing case study and numerical experiments, we quantify the computational trade-offs between these two formulations in terms of the solution optimality, computational time, and robustness against non-converging or trivial solutions.

math.OC↗

arXiv2025

Orbital Depot Location Optimization for Satellite Constellation Servicing with Low-Thrust Transfers

This paper addresses the critical problem of co-optimizing the optimal locations for orbital depots and the sequence of in-space servicing for a satellite constellation. While most traditional studies used network optimization for this problem, assuming a fixed set of discretized nodes in the network (i.e., a limited number of depot location candidates), this work is unique in that it develops a method to optimize the depot location in continuous space. The problem is formulated as mixed-integer nonlinear programming, and we propose a solution methodology that iteratively solves two decoupled problems: one using mixed-integer linear programming and the other using nonlinear programming with an analytic transfer solution. To demonstrate the effectiveness of our approach, we apply this methodology to a case study involving a GPS satellite constellation. Numerical experiments confirm the stability of our proposed solutions.

math.OC↗

arXiv2023

Rethinking Evaluation Metric for Probability Estimation Models Using Esports Data

Probability estimation models play an important role in various fields, such as weather forecasting, recommendation systems, and sports analysis. Among several models estimating probabilities, it is difficult to evaluate which model gives reliable probabilities since the ground-truth probabilities are not available. The win probability estimation model for esports, which calculates the win probability under a certain game state, is also one of the fields being actively studied in probability estimation. However, most of the previous works evaluated their models using accuracy, a metric that only can measure the performance of discrimination. In this work, we firstly investigate the Brier score and the Expected Calibration Error (ECE) as a replacement of accuracy used as a performance evaluation metric for win probability estimation models in esports field. Based on the analysis, we propose a novel metric called Balance score which is a simple yet effective metric in terms of six good properties that probability estimation metric should have. Under the general condition, we also found that the Balance score can be an effective approximation of the true expected calibration error which has been imperfectly approximated by ECE using the binning technique. Extensive evaluations using simulation studies and real game snapshot data demonstrate the promising potential to adopt the proposed metric not only for the win probability estimation model for esports but also for evaluating general probability estimation models.

cs.LG↗