AI for Detecting Approaches between Satellites and Debris A Scoping Review

Abstract

Goal: This scoping review aims to identify and characterize Artificial Intelligence (AI) techniques employed in detecting close approaches between satellites and space debris, with the objective of preventing in-orbit collisions. Design | Methodology |
Approach: The study was conducted following the Joanna Briggs Institute (JBI) methodology and reported in accordance with the PRISMA-ScR extension. The search strategy was applied to the Scopus and Web of Science databases, using predefined eligibility criteria and a dual-reviewer selection process. Results: The findings highlight the growing adoption of AI algorithms—such as machine learning and neural network—applied to orbital data analysis (TLEs and CDMs) to improve collision risk prediction accuracy. Originality | Value: The originality of this review lies in its focus on predictive techniques, offering a consolidated overview of existing approaches and identifying research gaps in the field of orbital safety.