Anton Kullberg

Background

Anton Kullberg defended his Ph.D. at the Division of Automatic Control at the Department of Electrical Engineering at Linköping University. His Ph.D. project was part of the Wallenberg AI, Autonomous Systems and Software Program (WASP). He received his M.Sc. degree in Applied Physics and Electrical Engineering (Y) in June 2019, also from Linköping University. He defended his Licentiate Thesis entitled Online Joint State Inference and Model Learning using Gaussian Process Approximations in December 2021. He defended his Ph.D. dissertation entitled Dynamic rEvolution: Adaptive state estimation via Gaussian processes and iterative filtering in June 2024.

Research

The topic of Anton's PhD project is to combine classical physical modeling with machine learning. The goal is to allow physical domain knowledge to be used in conjunction with learning of unknown and/or unmodeled behavior. A concrete example would be to use a simple motion model for a car and to learn unknown characteristics online, such as driver input/behavior.

Now, he primarily works with nonlinear filtering techniques and computationally efficient methods for Gaussian processes.