RoboDesign1M
IROS 2026 · 2026
RoboDesign1M introduces a large-scale dataset and foundation-model benchmark for robotic mechanism and design understanding.
The project focuses on multimodal robotic design data at scale, supporting downstream tasks such as retrieval, generation, and broader AI-assisted robot design workflows.
Paper: arXiv
Video
Publication
T. Le, T. Nguyen, Q. Tran, Q. Nguyen, B. Huang, H. Nguyen, M. N. Vu, T. D. Ta, and A. Nguyen, “RoboDesign1M: A Large-scale Dataset for Robot Design Understanding,” IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2026. arXiv PDF DOI.