Max Robotics — Global Service Robot Directory + US Certification

About

Matthias Seeger is a leading researcher at the intersection of machine learning and robotics, renowned for pioneering work in nonparametric regression for real-time control. His primary research areas include Gaussian process (GP) modeling, Bayesian inference, and robot inverse dynamics. Seeger’s major contribution is the development of Local Gaussian Process Regression (LGP), which enables fast, online model learning for high-performance robot control. His seminal 2009 paper, "Model Learning with Local Gaussian Process Regression" (322 citations), demonstrates how LGP can achieve precise torque models, allowing for more accurate, energy-efficient, and compliant robot control. This work, along with "Computed torque control with nonparametric regression models" (151 citations), addresses the critical challenge of unmodeled nonlinearities in rigid-body dynamics. Seeger also advanced semiparametric latent factor models (202 citations), offering efficient inference for multi-response regression. His research has had a profound impact on model-based control, with over 1,000 total citations, and his methods are widely adopted in robotics for real-time learning and control applications.