International Journal Of Imaging

Abstract

ML Based Predictive Models for Satellite Positioning and Velocity Estimation to Prevent Orbital Collision

Author: Sujoy Maji, T. Rahul Reddy , B. Prem Kumar , Dr. M. Thirupathi

Abstract:

The rapid growth in satellite deployments has significantly increased the congestion in Earth's orbital environment, leading to heightened risks of collisions. Traditionally, satellite positioning and velocity estimation have relied on physics-based models grounded in Keplerian mechanics and perturbation theories. While the methods have proven effective for many applications, their ability to adapt to real-time data and dynamic environmental factors such as atmospheric drag, solar radiation pressure, and gravitational perturbations remains limited. Moreover, traditional systems often struggle with scalability and accuracy when handling the increasing volume of satellite traffic and complex orbital scenarios. Machine learning (ML) presents a transformative approach to overcome the limitations by leveraging large datasets and adaptive algorithms to predict satellite trajectories with greater precision. By integrating historical telemetry data, environmental parameters, and advanced