MACHINE LEARNING-BASED POPULATION-LEVEL RISK RANKING OF SPACE OBJECTS USING ORBITAL AND DEBRIS-ENVIRONMENT FEATURES
AUTHORS:
Mentor
Affiliation
B.Tech 3rd Year CSE, SRM University, Andhra Pradesh
DOI: 
SHARE THIS ARTICLE
CC BY 4.0 License:
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
DOWNLOAD ARTICLE
REVIEW REPORT
CITE THIS ARTICLE
Abstract
The increasing number of cataloged resident space objects (RSOs) in low Earth orbit (LEO) has heightened concerns regarding the long-term sustainability of orbits and the potential for a self-perpetuating collisional cascade, known as the Kessler Syndrome. Existing machine learning approaches to space situational areness (SSA) are overwhelmingly reactive; they operate on pairwise Conjunction Data Messages (CDMs) issued only after a close approach has already been detected by a surveillance network, or they correct single-object orbit-propagation errors without evaluating the surrounding debris environment. In contrast, macro-scale stochastic and deterministic evolutionary models characterize population-level sustainability but cannot rank individual cataloged objects. This study addresses the resulting gap between object-level, CDM-dependent risk assessment and population-level, density-based environmental modeling by developing a pairwise-independent, physics-informed proxy-risk classification framework built entirely from standalone Two-Line Element (TLE) catalog data. A dataset of 34,136 currently orbiting payloads, debris fragments, and rocket bodies was constructed from the SATCAT and orbital element catalog, and twenty predictive features describing orbital geometry, local orbital crowding, altitude shell density, debris and rocket body neighbor density, and estimated relative velocity were engineered. A composite Overall Proxy-Risk Score, combining collision-risk and debris-cascade-risk proxies, was used to assign each object to a LOW, MEDIUM, or HIGH physics-informed proxy-risk category. Logistic Regression, Random Forest, and Gradient Boosting classifiers were evaluated using five-fold stratified cross-validation and an untouched 20% holdout test set (6,828 objects) against a majority-class dummy baseline. Logistic Regression achieved the highest holdout performance (accuracy = 0.9985, F1 = 0.9985, ROC-AUC = 0.9999), substantially exceeding the dummy baseline (accuracy = 0.4783). Feature importance analysis identified the relative velocity and local debris-neighbor density as dominant predictors. Because the proxy-risk label is a deterministic function of several of the input features, the very high classification performance is interpreted as evidence that the models recovered the structure of the proxy-risk formulation rather than as evidence of collision-prediction skill; this dependency is discussed explicitly as a central limitation. The framework is proposed as a lightweight, CDM-independent screening layer for prioritizing objects ahead of a detailed physics-based conjunction assessment, and not as a replacement for it.
Keywords
space debris space situational awareness orbital crowding machine learning proxy-risk classification Two-Line Elements Kessler Syndrome collision-risk screening Random Forest; Logistic Regression..
Article Information
PUBLICATION DETAILS
Review Type:
Double-Blind Peer Review
License:
CC BY-NC 4.0
Volume/Issue:
Volume 02, Issue 8
Review Rounds:
2
Article Type:
Research Article
Publication Date:
Aug 27 2026
Article Metrics
Article Views
27
PDF Downloads
2
HOW TO CITE
Copy
Varshini, J. G. (2026). Machine Learning-Based Population-Level Risk Ranking of Space Objects Using Orbital and Debris-Environment Features. International Journal of Science, Strategic Management and Technology, 02(8), 1-5. https://doi.org/10.55041/ijsmt.v2i8.069
Varshini, Juluru. "Machine Learning-Based Population-Level Risk Ranking of Space Objects Using Orbital and Debris-Environment Features." International Journal of Science, Strategic Management and Technology, vol. 02, no. 8, 2026, pp. 1-5. doi:https://doi.org/10.55041/ijsmt.v2i8.069.
Varshini, Juluru. "Machine Learning-Based Population-Level Risk Ranking of Space Objects Using Orbital and Debris-Environment Features." International Journal of Science, Strategic Management and Technology 02, no. 8 (2026): 1-5. https://doi.org/https://doi.org/10.55041/ijsmt.v2i8.069.
References
[1] D. J. Kessler and B. G. Cour-Palais, "Collision frequency of artificial satellites: The creation of a debris belt," J. Geophys. Res., vol. 83, no. A6, pp. 2637-2646, 1978.
[2] European Space Agency (ESA) Space Debris Office, "ESA Space Environment Report," ESA/ESOC, Darmstadt, Germany. [Online]. Available: https://www.sdo.esoc.esa.int/environment_report/Space_Environment_Report_latest.pdf
[3] T. S. Kelso, "Analysis of the Iridium 33-Cosmos 2251 collision," in Proc. Advanced Maui Optical and Space Surveillance Technologies Conf. (AMOS), Maui, HI, USA, 2009.
[4] Inter-Agency Space Debris Coordination Committee (IADC), "IADC Space Debris Mitigation Guidelines," IADC-02-01, Rev. 2, Mar. 2020.
[5] D. A. Vallado, P. Crawford, R. Hujsak, and T. S. Kelso, "Revisiting Spacetrack Report #3," in Proc. AIAA/AAS Astrodynamics Specialist Conf., Keystone, CO, USA, Aug. 2006, AIAA 2006-6753.
[6] B. Li, Y. Zhang, J. Huang, and J. Sang, "Improved orbit predictions using two-line elements through error pattern mining and transferring," Acta Astronaut., vol. 188, pp. 405-415, 2021.
[7] M. R. Akella and K. T. Alfriend, "Probability of collision between space objects," J. Guid., Control, Dyn., vol. 23, no. 5, pp. 769-772, 2000.
[8] R. P. Patera, "General method for calculating satellite collision probability," J. Guid., Control, Dyn., vol. 24, no. 4, pp. 716-722, 2001.
[9] F. K. Chan, Spacecraft Collision Probability. El Segundo, CA, USA: Aerospace Press, 2008.
[10] T. Uriot, D. Izzo, L. F. Simoes, R. Abay, N. Einecke, S. Rebhan, J. Martinez-Heras, F. Letizia, J. Siminski, and K. Merz, "Spacecraft collision avoidance challenge: Design and results of a machine learning competition," Astrodynamics, vol. 6, pp. 121-140, 2022.
[11] L. Tulczyjew, M. Myller, M. Kawulok, D. Kostrzewa, and J. Nalepa, "Predicting risk of satellite collisions using machine learning," J. Space Safety Eng., vol. 8, no. 4, pp. 339-344, 2021.
[12] N. Boscolo Fiore, "Machine learning based satellite collision avoidance strategy," M.S. thesis, Dept. Aerosp. Sci. Technol., Politecnico di Milano, Milan, Italy, 2021.
[13] S. Metz, H. Simon, and F. Letizia, "Implementation and comparison of data-based methods for collision avoidance in satellite operations," in Proc. 8th European Conf. Space Debris (ECSD), Darmstadt, Germany, 2021.
[14] S. Nikolaev, D. Phillion, H. K. Springer, W. deVries, M. Jiang, A. Pertica, J. Henderson, M. Horsley, and S. Olivier, "Brute force modeling of the Kessler syndrome," in Proc. Advanced Maui Optical and Space Surveillance Technologies Conf. (AMOS), Maui, HI, USA, 2012.
[15] J. Hudson, "KESSYM: A stochastic orbital debris model for evaluation of Kessler Syndrome risks and mitigations," J. Student Res., 2023.
[16] N. L. Johnson, P. H. Krisko, J.-C. Liou, and P. D. Anz-Meador, "NASA's new breakup model of EVOLVE 4.0," Adv. Space Res., vol. 28, no. 9, pp. 1377-1384, 2001.
[17] L. Sanchez, M. Vasile, S. Sanvido, K. Merz, and C. Taillan, "On the use of machine learning and evidence theory to improve collision risk management," Acta Astronaut., vol. 181, pp. 694-706, 2021.
[2] European Space Agency (ESA) Space Debris Office, "ESA Space Environment Report," ESA/ESOC, Darmstadt, Germany. [Online]. Available: https://www.sdo.esoc.esa.int/environment_report/Space_Environment_Report_latest.pdf
[3] T. S. Kelso, "Analysis of the Iridium 33-Cosmos 2251 collision," in Proc. Advanced Maui Optical and Space Surveillance Technologies Conf. (AMOS), Maui, HI, USA, 2009.
[4] Inter-Agency Space Debris Coordination Committee (IADC), "IADC Space Debris Mitigation Guidelines," IADC-02-01, Rev. 2, Mar. 2020.
[5] D. A. Vallado, P. Crawford, R. Hujsak, and T. S. Kelso, "Revisiting Spacetrack Report #3," in Proc. AIAA/AAS Astrodynamics Specialist Conf., Keystone, CO, USA, Aug. 2006, AIAA 2006-6753.
[6] B. Li, Y. Zhang, J. Huang, and J. Sang, "Improved orbit predictions using two-line elements through error pattern mining and transferring," Acta Astronaut., vol. 188, pp. 405-415, 2021.
[7] M. R. Akella and K. T. Alfriend, "Probability of collision between space objects," J. Guid., Control, Dyn., vol. 23, no. 5, pp. 769-772, 2000.
[8] R. P. Patera, "General method for calculating satellite collision probability," J. Guid., Control, Dyn., vol. 24, no. 4, pp. 716-722, 2001.
[9] F. K. Chan, Spacecraft Collision Probability. El Segundo, CA, USA: Aerospace Press, 2008.
[10] T. Uriot, D. Izzo, L. F. Simoes, R. Abay, N. Einecke, S. Rebhan, J. Martinez-Heras, F. Letizia, J. Siminski, and K. Merz, "Spacecraft collision avoidance challenge: Design and results of a machine learning competition," Astrodynamics, vol. 6, pp. 121-140, 2022.
[11] L. Tulczyjew, M. Myller, M. Kawulok, D. Kostrzewa, and J. Nalepa, "Predicting risk of satellite collisions using machine learning," J. Space Safety Eng., vol. 8, no. 4, pp. 339-344, 2021.
[12] N. Boscolo Fiore, "Machine learning based satellite collision avoidance strategy," M.S. thesis, Dept. Aerosp. Sci. Technol., Politecnico di Milano, Milan, Italy, 2021.
[13] S. Metz, H. Simon, and F. Letizia, "Implementation and comparison of data-based methods for collision avoidance in satellite operations," in Proc. 8th European Conf. Space Debris (ECSD), Darmstadt, Germany, 2021.
[14] S. Nikolaev, D. Phillion, H. K. Springer, W. deVries, M. Jiang, A. Pertica, J. Henderson, M. Horsley, and S. Olivier, "Brute force modeling of the Kessler syndrome," in Proc. Advanced Maui Optical and Space Surveillance Technologies Conf. (AMOS), Maui, HI, USA, 2012.
[15] J. Hudson, "KESSYM: A stochastic orbital debris model for evaluation of Kessler Syndrome risks and mitigations," J. Student Res., 2023.
[16] N. L. Johnson, P. H. Krisko, J.-C. Liou, and P. D. Anz-Meador, "NASA's new breakup model of EVOLVE 4.0," Adv. Space Res., vol. 28, no. 9, pp. 1377-1384, 2001.
[17] L. Sanchez, M. Vasile, S. Sanvido, K. Merz, and C. Taillan, "On the use of machine learning and evidence theory to improve collision risk management," Acta Astronaut., vol. 181, pp. 694-706, 2021.
Ethics and Compliance
✓ All ethical standards met
This article has undergone plagiarism screening and double-blind peer review. Editorial policies have been followed. Authors retain copyright under CC BY-NC 4.0 license. The research complies with ethical standards and institutional guidelines.
Indexed In
Similar Articles
AI-GCF: An Artificial Intelligence-Based Green Communication Framework for Energy-Efficient and Sustainable Communication Networks
(2026)
DOI: 10.55041/ijsmt.v2i7.027
AI Based Driver Drowsiness & Distraction Detection
(2026)
DOI: 10.55041/ijsmt.v2i4.436
A Secure Voting System using Blockchain Technology
(2026)
DOI: 10.55041/ijsmt.v2i5.256
Previous Article
