Mobina Mobaraki

University of British Columbia, Shahid Beheshti University

Papers

Total Citations

H-Index

About

Mobina Mobaraki is a researcher at the forefront of intelligent manufacturing, specializing in the integration of deep learning and computer vision for automated welding and robotic control. Her primary contributions lie in developing vision-based systems that enable welding robots to adapt to real-world uncertainties, significantly improving efficiency and quality in small and medium batch production. Her most cited work, "Vision-based seam tracking for GMAW fillet welding based on keypoint detection deep learning model" (2024, 17 citations), introduces a novel approach that allows robots to dynamically track weld seams, overcoming the limitations of pre-programmed systems in noisy environments. Mobaraki has also advanced defect classification in gas metal arc welding by leveraging multi-modal data, as demonstrated in her 2025 paper (2 citations), which enhances real-time quality control. Earlier, she explored control systems in robotics with her work on PID controller robustness for Cartesian robots (2019, 3 citations). Her research bridges the gap between theoretical deep learning models and practical industrial applications, offering scalable solutions for adaptive automation. With a growing citation record, Mobaraki is establishing herself as a key innovator in smart welding and robotic perception.