Puze Liu

Research Interests
Robotics, Reinforcement Learning, Control and Optimization, Safe Reinforcement Learning.
Affiliations
- TU Darmstadt, Intelligent Autonomous Systems, Computer Science Department
- German Research Center for AI (DFKI), Research Department: SAIROL
Contact
Contact
puze.liu(at)dfki(dot)de
puze.liu@robot-learning.de
Room 2.1.16, Building S4|14, DFKI, SAIROL, Mornewegstraße 30, 64293 Darmstadt
Puze joined Intelligent Autonomous Systems in Aug 2019 as a Ph.D. student. He work with Davide Tateo and Jan Peters within the research group Safe and Reliable Robot Learning. His research interests include robotics, reinforcement learning, controlling, and human-robot collaboration. Puze is looking for new approaches for robots to learn collaborative tasks that enable robots to assist humans in different scenarios.
Before starting his Ph.D., Puze completed his Master's degree in Computational Engineering Science at the Technische Universitaet Berlin. During his study, Puze's previous work focused on robot fundamental algorithms and safety problems during human-robot interaction. His master thesis "Industrial Robot Motion Control depending on the Distance between Human and Robot" was under the supervision of Prof. Dr-Ing. Jörg Krüger and Martin J. Rosenstrauch.
Key References
-
- Liu, P.; Bou-Ammar H.; Peters, J.; Tateo D. (conditionally accepted). Safe Reinforcement Learning on the Constraint Manifold: Theory and Applications, Submitted to the IEEE Transactions on Robotics (T-Ro).
-
- Liu, P.; Zhang, K.; Tateo, D.; Jauhri, S.; Hu, Z.; Peters, J. Chalvatzaki, G. (2023). Safe Reinforcement Learning of Dynamic High-Dimensional Robotic Tasks: Navigation, Manipulation, Interaction, 2023 IEEE International Conference on Robotics and Automation (ICRA), IEEE.
-
- Liu, P.; Zhang, K.; Tateo, D.; Jauhri, S.; Peters, J.; Chalvatzaki, G.; (2022). Regularized Deep Signed Distance Fields for Reactive Motion Generation, 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).
-
--- Best Paper Award Finalist ---- Liu, P.; Tateo, D.; Bou-Ammar, H.; Peters, J. (2021). Robot Reinforcement Learning on the Constraint Manifold, Proceedings of the Conference on Robot Learning (CoRL).
-
--- Best Entertainment and Amusement Paper Award Finalist ---- Liu, P.; Tateo, D.; Bou-Ammar, H.; Peters, J. (2021). Efficient and Reactive Planning for High Speed Robot Air Hockey, Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).
-
- Guenster, J.; Liu, P.; Peters, J.; Tateo, D. (2024). Handling Long-Term Safety and Uncertainty in Safe Reinforcement Learning, Proceedings of the Conference on Robot Learning (CoRL).
-
- Kicki, P.; Tateo; D., Liu, P.; Guenster, J.; Peters, J.; Walas, K. (2024). Bridging the gap between Learning-to-plan, Motion Primitives and Safe Reinforcement Learning, Proceedings of the Conference on Robot Learning (CoRL).
-
- Gu, S.; Liu, P.; Kshirsagar, A.; Chen, G.; Peters, J.; Knoll, A. (2024). ROSCOM: Robust Safe Reinforcement Learning on Stochastic Constraint Manifolds, IEEE Transactions on Automation Science and Engineering (T-ASE).
-
- Kicki, P.; Liu, P.; Tateo, D.; Bou Ammar, H.; Walas, K.; Skrzypczynski, P.; Peters, J. (2024). Fast Kinodynamic Planning on the Constraint Manifold with Deep Neural Networks, IEEE Transactions on Robotics (T-Ro), and Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 40, pp.277-297.
-
- Memmel, M.; Liu, P.; Tateo, D.; Peters, J. (2022). Dimensionality Reduction and Prioritized Exploration for Policy Search, 25th International Conference on Artificial Intelligence and Statistics (AISTATS).
-
- 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).
- 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).
Supervision: Completed Theses and Projects
Start | Type | In coorperation with | Student(s) | Topic | Document |
---|---|---|---|---|---|
2024 | Master Thesis | Davide Tateo | Jonas Günster | Handling Long-Term Safety and Uncertainty in Safe Reinforcement Learning | |
2024 | Master Thesis | Davide Tateo | Yuheng Ouyang | Hierarchical Reinforcement Learning with Self-Play for Robotic Air Hockey | |
2023 | Master Thesis | Davide Tateo | Johannes Heeg | Task Space Exploration in Robot Reinforcement Learning | |
2022 | Master Thesis | Davide Tateo, Georgia Chalvatzaki | Kuo Zhang | Learning Geometric Constraints for Safe Robot Interactions | |
2021 | Master Thesis | Davide Tateo | Verena Sieburgur | Development of a Baseline Agent in Robot Air Hockey | |
2020 | Master Thesis | - | Jianhang He | Imitation Learning with Energy Based Model | |
2020 | Master Thesis | Julen Urain | Zhenhui Zhou | Approximated Policy Search in Black-Box Optimization | |
2022 | Bachelor Thesis | Davide Tateo | Jonas Günster | Learning the Low Level Policy for Robot Air Hockey | |
2024 | Project | Theo Gruner | Shihao Li, Yu Wang | Benchmarking Sequence Models for Discontinuous Dynamical System | |
2023 | Project | Theo Gruner | Anna Klyushina, Marcel Rath | Black Box System Identification for an Air-Hockey Environment | |
2023/2024 | Project | Piotr Kicki | Niclas Merten | Kinodynamic Neural Planner for low-level Robot Air Hockey Hitting | |
2023 | Project | Davide Tateo | Yuheng Ouyang | Learning High-Level Policy for Robot Air Hockey using DQN | |
2022 | Project | - | Haozhe Zhu | On the Improvement of the Baseline Agent for Robot Air Hockey | |
2022 | Project | Davide Tateo | Jonas Günster | Building Up Resetting Device for Robot Air Hockey using PA10 | |
2022 | Project | - | Haoran Ding | Learning the Residual Dynamics using Extended Kalman Filter for Puck Tracking | |
2022 | Project | Davide Tateo | Andreas Seidl Fernandez, Joshua Johannson | Legged Locomotion for Quadruped Robots | |
2020 | Project | Davide Tateo | Verena Sieburgur, Chen Xue | Bayesian Optimization for System Identification in Robot Air Hockey | |
2020 | Project | Julen Urain | Niklas Babendererde, Jiawei Huang | Benchmarking Multi-Arm Bandit & Black Box optimization(DFO) 4 Grasping | |
2019 | Project | Davide Tateo | Patrick Lutz | Robot Air-Hockey |