Michael Mathew

University of Birmingham, Central Mechanical Engineering Research Institute

Papers

Total Citations

H-Index

About

Michael Mathew’s research focuses on advancing robot autonomy through learning-based approaches, particularly in manipulation and self-modeling. His major contributions lie in developing robust planning methods for robot manipulation under uncertainty. In his highly cited work, “Uncertainty averse pushing with model predictive path integral control” (2017, 39 citations), Mathew demonstrates how to plan robust push manipulations by learning forward models from robot data, using techniques like Gaussian Process Regression and ensemble methods. This work addresses the critical challenge of enabling robots to operate effectively despite imperfect models and environmental uncertainty. Mathew also contributed to the field of robot self-modeling in his 2014 paper, exploring how robots can learn their own dynamics from data rather than relying solely on designer-calculated models, which is essential for adaptive and resilient robotic systems. His research bridges machine learning and control, offering practical solutions for real-world robot manipulation. With a focus on data-driven methods and uncertainty-aware planning, Mathew’s work has direct implications for industrial automation, service robotics, and autonomous systems, making him a notable figure in modern robotics research.