Michael U. Gutmann

Papers

Total Citations

H-Index

About

Michael U. Gutmann is a leading researcher at the intersection of robotics, machine learning, and physical reasoning, with a focus on enabling robots to manipulate complex, real-world materials. His key contributions lie in developing algorithms that allow robots to learn and adapt their actions through approximate physical models and efficient calibration. Gutmann’s work on adaptable pouring, for instance, teaches robots to handle liquids without spilling by using fast but approximate fluid simulations, a practical approach that has garnered 29 citations. He further advanced this line of research with "Stir to Pour," a method that mimics human probing actions—like tilting a bottle—to efficiently calibrate liquid properties for more robust pouring strategies, earning 15 citations. Beyond his technical papers, Gutmann played a pivotal role in organizing the 1st Annual Conference on Robot Learning (CoRL 2017), a landmark event that helped shape the field and has been cited 14 times. His work is notable for bridging simulation and real-world physics, offering scalable solutions for robotic manipulation in everyday environments. With a growing impact, Gutmann’s research continues to inspire new approaches in robot learning and adaptive control.