Project Description
We aim to develop and integrate AI foundation models of various modalities such as language, images, acoustics, and 3D point clouds to build robotics foundation models that enables motion planning and control of robots in the real world. The CNRS-AIST JRL, in collaboration with the Industrial CPS Research Center and the Artificial Intelligence Research Center, is working to develop robotics foundation models through the development of imitation learning methods that can realize difficult manipulation tasks in a data-driven manner.
PI in JRL
Masaki MurookaPeriod
2024.4 - 2027.3Funded by
AISTRelated Publications (including JRL contributions)
| Title | Authors | Conference/Book | Year | bib | mov | prj | |
| RoboManipBaselines: A Unified Framework for Imitation Learning in Robotic Manipulation across Real and Simulation Environments | M. Murooka, T. Motoda, R. Nakajo, H. Oh, K. Makihara, K. Shirai, T. Ogata, Y. Domae | IEEE Access | 2026 | ||||
| Self-augmented robot trajectory: efficient imitation learning via safe self-augmentation with demonstrator-annotated precision | H. Oh, M. Murooka, T. Motoda, R. Nakajo, Y. Domae | Advanced Robotics | 2026 | ||||
| Learning Bimanual Manipulation Via Action Chunking and Inter-Arm Coordination with Transformers | T. Motoda, R. Hanai, R. Nakajo, M. Murooka, F. Erich, Y. Domae | IEEE International Conference on Automation Science and Engineering | 2025 |