Aditya Bhatt
Research Interests
Soft Robotics, Dexterous Manipulation, Model-Predictive Control, Reinforcement Learning
Affiliations
1. TU Darmstadt, Intelligent Autonomous Systems, Computer Science Department
2. German Research Center for AI (DFKI), Research Department: SAIROL
Contact
aditya.bhatt@dfki.de
Room 2.1.16, Building S4|14, DFKI, SAIROL, Mornewegstraße 30, 64293 Darmstadt
Aditya Bhatt joined the Intelligent Autonomous Systems Group in November 2022. He is employed as a researcher at DFKI, supervised by Boris Belousov and Jan Peters. Having previously designed dexterous manipulation strategies with soft robot hands, he now aims to exploit the principles of compliance and model-free feedback control as inductive biases for learning manipulation skills. His goal is to get robots to learn dexterous manipulation skills efficiently from as little data as possible.
Publications
- Bhatt, A.; Palenicek, D.; Belousov, B.; Argus, M.; Amiranashvili, A.; Brox, T.; Peters, J. (2024). CrossQ: Batch Normalization in Deep Reinforcement Learning for Greater Sample Efficiency and Simplicity, International Conference on Learning Representations (ICLR), Spotlight.
- Bhatt, A.; Palenicek, D.; Belousov, B.; Argus, M.; Amiranashvili, A.; Brox, T.; Peters, J. (2024). CrossQ: Batch Normalization in Deep Reinforcement Learning for Greater Sample Efficiency and Simplicity, European Workshop on Reinforcement Learning (EWRL).
- Meser, M.; Bhatt, A.; Belousov, B.; Peters, J. (2024). MuJoCo MPC for Humanoid Control: Evaluation on HumanoidBench, 40th Anniversary of the IEEE International Conference on Robotics and Automation (ICRA@40).
- Bhatt, A.; Sieler, S.; Puhlmann, S.; Brock, O. (2021). Surprisingly Robust In-Hand Manipulation: An Empirical Study, Robotics: Science and Systems.