Julen Urain De Jesus

Julen joined the Intelligent Autonomous Systems Group at TU Darmstadt as a Ph.D. researcher in January 2019. He received his MS in Automatic Control and Robotics from UPC (Barcelona) in November 2017. Julen has gained a lot of knowledge in machine learning and robotics from different institutions. He developed his master's thesis in BioRob lab in EPFL(Laussane) under the supervision of Auke Ijspeert and Jessica Lanini. He did an internship in the first edition of Deep Learning and Robotics Challenge (DLRC) in VW Data:Lab (Munich) and for the last year from December 2017 to December 2018 He has worked as a robotics researcher in IK4-Tekniker (Eibar).

Julen was selected as an R:SS Pioneer for robotics research in 2023 and as an intern at Nvidia Robotics in 2022-23.

Julen's main research line is in the integration of machine learning algorithms to learn complex manipulation skills. During the Ph.D., he is looking for new ways of representing the dynamics and policies, following probabilistic models. It is expected that more structured models for representing both the world and the policy will improve the performance of the robot in terms of generalization and adaptation. In his thesis, Julen studies the integration of Generative Models (Energy-Based Models (EBM), Normalizing Flows, Diffusion Models) and Optimization with Robotics elements (Stability, Differential Geometry, Control) to learn complex manipulation skills and improve the generalization capabilities.

Key References

Cost Learning

1. •       Bib
     Urain, J.; Funk, N.; Peters, J.; Chalvatzaki G (2023). SE(3)-DiffusionFields: Learning smooth cost functions for joint grasp and motion optimization through diffusion, International Conference on Robotics and Automation (ICRA).
2. •       Bib
     Urain, J.; Le, A. T.; Lambert, A.; Chalvatzaki, G.; Boots, B.; Peters, J. (2022). Learning Implicit Priors for Motion Optimization, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).
   •   Bib
     Le, A. T.; Urain, J.; Lambert, A.; Chalvatzaki, G.; Boots, B.; Peters, J. (2022). Learning Implicit Priors for Motion Optimization, RSS 2022 Workshop on Implicit Representations for Robotic Manipulation.

Motion Optimization with Energy-Based Models

1. •     Bib
     Urain, J.; Li, A.; Liu, P.; D'Eramo, C.; Peters, J. (2023). Composable energy policies for reactive motion generation and reinforcement learning, International Journal of Robotics Research (IJRR).
   •     Bib
     Urain, J.; Li, A.; Liu, P.; D'Eramo, C.; Peters, J. (2021). Composable Energy Policies for Reactive Motion Generation and Reinforcement Learning, Robotics: Science and Systems (RSS).
2. •       Bib
     Urain, J.; Le, A. T.; Lambert, A.; Chalvatzaki, G.; Boots, B.; Peters, J. (2022). Learning Implicit Priors for Motion Optimization, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).
   •   Bib
     Le, A. T.; Urain, J.; Lambert, A.; Chalvatzaki, G.; Boots, B.; Peters, J. (2022). Learning Implicit Priors for Motion Optimization, RSS 2022 Workshop on Implicit Representations for Robotic Manipulation.

Motion Primitives and Dynamic Systems

1. •     Bib
     Urain, J.; Tateo, D.; Peters, J. (2023). Learning Stable Vector Fields on Lie Groups, Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), IEEE R-AL Track.
   •     Bib
     Urain, J.; Tateo, D; Peters, J. (2022). Learning Stable Vector Fields on Lie Groups, Robotics and Automation Letters (RA-L).
2. •     Bib
     Urain, J.; Ginesi, M.; Tateo, D.; Peters, J. (2020). ImitationFlow: Learning Deep Stable Stochastic Dynamic Systems by Normalizing Flows, IEEE/RSJ International Conference on Intelligent Robots and Systems.