Mahmoud Shoman

Papers

Total Citations

H-Index

About

Mahmoud Shoman’s research lies at the intersection of robotics, machine learning, and human-robot interaction, with a specific focus on **trajectory learning from demonstration**. His work addresses a fundamental challenge in modern robotics: how to efficiently teach robots to perform complex tasks by observing human demonstrations, rather than through explicit programming. Shoman’s major contributions include pioneering the use of **Principal Component Analysis (PCA)** for trajectory learning, demonstrating how dimensionality reduction can extract essential movement patterns from demonstration data. He further advanced the field by introducing **posterior hidden Markov model state distributions** to capture temporal dependencies in learned trajectories, enabling robots to generalize from limited examples. His comparative study of preprocessing techniques for trajectory learning provides a systematic framework for selecting optimal methods across different robotic applications. While his citation counts remain modest—with his most cited work reaching 5 citations—his research represents foundational steps toward more intuitive robot programming. Shoman’s work is particularly relevant for students and researchers exploring **learning from demonstration**, **robot skill acquisition**, and **human-robot collaboration**, offering practical methodologies for developing robots that can learn naturally from human teachers.