Welcome to the Geometric Robot Learning Lab!
In the GeoRob Lab, we develop data-efficient robot learning, optimization, and control algorithms with sound theoretical guarantees. To do so, we follow an interdisciplinary approach that considers inductive bias, typically in the form of geometry and physics, as cornerstone.
PhD students
Loizos Hadjiloizou
Geometry and symmetries of operational space learning and control
Riccardo Morandi
Geometries of deformable object manipulation
Katharina Friedl
Physics-informed geometric learning and control for dynamical systems
Main supervisor: Danica Kragic
Postdocs
Generalizable imitation learning via inductive biases
Vision-based geometric robot learning for manipulation
External PhD students
Research Engineers
Ulises Campodónico
Martín Gallegos
Alumnis
Federico Pavesi
Bayesian optimization on Riemannian manifolds
Visiting PhD student, University of Milano-Bicocca