Mark Sastuba

TU Bergakademie Freiberg, Federal Railway Authority

Papers

Total Citations

155

H-Index

About

Mark Sastuba is a leading researcher in the intersection of robotics and machine learning, with a primary focus on enhancing physical human-robot interaction. His work centers on developing data-driven methods for robots to detect, estimate, and react to external perturbations—unexpected forces or movements introduced by human partners during collaborative tasks. Sastuba’s major contribution is pioneering the use of Dynamic Mode Decomposition (DMD) for perturbation estimation, a novel approach that allows robots to sense and compensate for external forces without relying on expensive, specialized force sensors. His seminal 2015 paper on this topic has garnered 124 citations, establishing a foundational technique in the field. His earlier 2014 work (27 citations) further solidified this methodology. More recently, Sastuba has expanded into sensor registration, addressing challenges in converting depth images and point clouds for feature-based pose estimation (2023). By enabling more robust, sensor-efficient robotic behavior, his research directly advances the safety and fluidity of collaborative robots, making them more practical for real-world human environments.