IEEE/RSJ IROS WS — Geometric Representations
in Robotics
Structured priors, metrics, constraints &
groundings for Manipulation
Invited Speakers
We are delighted to have a distinguished line-up of speakers from across sensing, learning, planning, and control — united by the role of geometric structure in robotic manipulation.
Georgia Chalvatzaki leads the PEARL Lab at TU Darmstadt. Her research lies at the intersection of robot learning, planning under uncertainty, and structured, geometrically-grounded representations for manipulation.
Arne Sachtler's research focuses on manifold-valued motor skills, Riemannian geometry for robot learning, and structure-preserving representations for manipulation in bimanual and dexterous tasks.
Constantinos Chamzas works on learning-guided motion planning and geometric representations for robot manipulation, with emphasis on Experience-based Planning Domains and structured latent spaces.
Jonathan Kelly leads the STARS Laboratory at U of Toronto. His research focuses on differential geometric methods in estimation and planning, robust computer vision, and collaborative mobile manipulation.
Katerina Fragkiadaki's research lies at the intersection of computer vision, machine learning, and robotics, focusing on predictive world models to reason through physical dynamics and sim2real robot learning.
Johannes Lachner's research focuses on geometric methods for robot control and manipulation, including task-space control on Lie groups, geometric fabrics, and structured optimization.
Tobias Löw's research focuses on geometric methods and learning for robotic manipulation and motion generation, with application to complex constrained environments.