Contact
Dr.-Ing. Jens Kober
Associate Professor, Cognitive Robotics department (CoR)
Delft University of Technology (TU Delft)
Mekelweg 2 (ME Building)
2628 CD Delft, The Netherlands
🖂 j.kober (at) tudelft.nl
🕿 +31 (0)15 27 85150
Office: F-2-080
https://www.linkedin.com/in/jens-kober/
Research interests
Robotics, Motor Skill Learning, Machine Learning, Reinforcement Learning, Imitation Learning, Deep Learning, Interactive Learning
News
- 4 new preprints:
. Beyond Behavior Cloning: Robustness through Interactive Imitation and Contrastive Learning. arXiv:2502.07645 [cs.RO], 2025. [bibtex] [pdf] [webpage] [doi]
. Versatile, Robust, and Explosive Locomotion with Rigid and Articulated Compliant Quadrupeds. arXiv:2504.12854 [cs.RO], 2025. [bibtex] [pdf] [doi] [video]
. MUKCa: Accurate and Affordable Cobot Calibration Without External Measurement Devices. arXiv:2503.12584 [cs.RO], 2025. [bibtex] [pdf] [doi] [code] [video]
. Explosive Jumping with Rigid and Articulated Soft Quadrupeds via Example Guided Reinforcement Learning. arXiv:2503.16197 [cs.RO], 2025. [bibtex] [pdf] [doi] [video]
- Jens Kober is on the TU Delft front page with a ‘double-interview’ on “Bringing machine learning into the real world”
- 4 accepted conference papers:
. ExploRLLM: Guiding Exploration in Reinforcement Learning with Large Language Models. In IEEE International Conference on Robotics and Automation (ICRA), 2025. Accepted. [bibtex] [pdf] [webpage]
. Output-Constrained Individual Pitch Control using an Adaptive Leaky Integrator for Wind Turbine Blade Load Reductions. In American Control Conference (ACC), 2025. Accepted.
. Noise-conditioned Energy-based Annealed Rewards (NEAR): A Generative Framework for Imitation Learning from Observation. In 13th International Conference on Learning Representations (ICLR), 2025. Jens Kober and Jan Peters supervised equally. [bibtex] [pdf] [webpage] [code]
. Impedance Primitive-augmented Hierarchical Reinforcement Learning for Sequential Tasks. In IEEE International Conference on Robotics and Automation (ICRA), 2025. Accepted.
- 1 accepted journal paper:
. REX: GPU-Accelerated Sim2Real Framework with Delay and Dynamics Estimation. Transactions on Machine Learning Research, 2025. [bibtex] [pdf] [url] [code] [video]
- 1 new preprint:
. Mode Collapse Happens: Evaluating Critical Interactions in Joint Trajectory Prediction Models. TechRxiv, 2025. [bibtex] [pdf] [doi]
- Jens Kober is co-organizing an IEEE RAM Special Issue on Embodied AI: Bridging Robotics and Artificial Intelligence Toward Real-World Applications
- 4 new preprints:
. ILeSiA: Interactive Learning of Situational Awareness from Camera Input. arXiv:2409.20173 [cs.RO], 2024. [bibtex] [pdf] [webpage] [doi] [code] [video]
. Task-unaware Lifelong Robot Learning with Retrieval-based Weighted Local Adaptation. arXiv:2410.02995 [cs.RO], 2024. [bibtex] [pdf] [doi]
. Mastering Contact-rich Tasks by Combining Soft and Rigid Robotics with Imitation Learning. arXiv:2410.07787 [cs.RO], 2024. [bibtex] [pdf] [doi]
. Learning Adaptive Hydrodynamic Models Using Neural ODEs in Complex Conditions. arXiv:2410.00490 [cs.RO], 2024. [bibtex] [pdf] [doi]
- Accepted CoRL Demo: A Hybrid Soft-Rigid Robotic Platform Showcasing Learning of Generalizable Skills from Demonstration - Mariano Ramírez Montero, Ebrahim Shahabi, Giovanni Franzese, Jens Kober, Barbara Mazzolai, Cosimo Della Santina
- 1 new preprint:
. Leveraging LLMs, Graphs and Object Hierarchies for Task Planning in Large-Scale Environments. arXiv:2409.04775 [cs.RO], 2024. [bibtex] [pdf] [doi] [video]
- Outstanding Paper Award Empirical Resourcefulness in RL:
. An Open-Loop Baseline for Reinforcement Learning Locomotion Tasks. Reinforcement Learning Journal, 1(1), 2024. Reinforcement Learning Conference (RLC). [bibtex] [pdf] [html] [code] [video]