Predicting RCS SIZE of Space Debris

Project Overview:

• Hypothesis Testing:Utilized K^2 normality test and correlation analysis to understand the distribution and relationships of key features, such as RCS_SIZE and OBJECT_TYPE.
• Machine Learning Algorithms: Explored various machine learning algorithms, including Logistic Regression, Gaussian Process Classifier, Support Vector Machine, and others, to build predictive models for RCS size estimation.
• Performance Evaluation: Evaluated the performance of each algorithm, achieved high accuracy scores ranging from 90% to 94% and compared their effectiveness in predicting space debris RCS size, providing insights into model selection and feature importance.

Future Directions: While the project achieved promising results, future work could involve further refinement of existing models, exploration of additional features, and integration of real-time data for more robust predictions. Additionally, expanding the scope to include other properties of space debris could offer a more comprehensive understanding of its behavior.

This project demonstrated my ability to apply advanced statistical techniques and machine learning algorithms to tackle complex problems in the aerospace domain. By leveraging data-driven approaches, I contributed to the advancement of space debris prediction, emphasizing the importance of data analysis and modeling in aerospace engineering.