Fabrica: Dual-Arm Assembly of General Multi-Part Objects via Integrated Planning and Learning

Yunsheng Tian, Joshua Jacob, Yijiang Huang, Jialiang Zhao, Edward Li Gu, Pingchuan Ma, Annan Zhang, Farhad Javid, Branden Romero, Sachin Chitta, Shinjiro Sueda, Hui Li, Wojciech Matusik
Proceedings of The 9th Conference on Robot Learning, PMLR 305:406-419, 2025.

Abstract

Multi-part assembly poses significant challenges for robotic systems to execute long-horizon, contact-rich manipulation with generalization across complex geometries. We present a dual-arm robotic system capable of end-to-end planning and control for autonomous assembly of general multi-part objects. For planning over long horizons, we develop hierarchies of precedence, sequence, grasp, and motion planning with automated fixture generation, enabling general multi-step assembly on any dual-arm robots. The planner is made efficient through a parallelizable design and is optimized for downstream control stability. For contact-rich assembly steps, we propose a lightweight reinforcement learning framework that trains generalist policies across object geometries, assembly directions, and grasp poses, guided by equivaraiance and residual actions obtained from the plan. These policies transfer zero-shot to the real world and achieve 80% success rates. For systematic evaluation, we propose a benchmark suite of multi-part assemblies resembling industrial and daily objects across diverse categories and geometries. By integrating efficient global planning and robust local control, we demonstrate the first system to achieve complete and generalizable real-world multi-part assembly without domain knowledge or human demonstrations.

Cite this Paper

BibTeX
@InProceedings{pmlr-v305-tian25a, title = {Fabrica: Dual-Arm Assembly of General Multi-Part Objects via Integrated Planning and Learning}, author = {Tian, Yunsheng and Jacob, Joshua and Huang, Yijiang and Zhao, Jialiang and Gu, Edward Li and Ma, Pingchuan and Zhang, Annan and Javid, Farhad and Romero, Branden and Chitta, Sachin and Sueda, Shinjiro and Li, Hui and Matusik, Wojciech}, booktitle = {Proceedings of The 9th Conference on Robot Learning}, pages = {406--419}, year = {2025}, editor = {Lim, Joseph and Song, Shuran and Park, Hae-Won}, volume = {305}, series = {Proceedings of Machine Learning Research}, month = {27--30 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v305/main/assets/tian25a/tian25a.pdf}, url = {https://proceedings.mlr.press/v305/tian25a.html}, abstract = {Multi-part assembly poses significant challenges for robotic systems to execute long-horizon, contact-rich manipulation with generalization across complex geometries. We present a dual-arm robotic system capable of end-to-end planning and control for autonomous assembly of general multi-part objects. For planning over long horizons, we develop hierarchies of precedence, sequence, grasp, and motion planning with automated fixture generation, enabling general multi-step assembly on any dual-arm robots. The planner is made efficient through a parallelizable design and is optimized for downstream control stability. For contact-rich assembly steps, we propose a lightweight reinforcement learning framework that trains generalist policies across object geometries, assembly directions, and grasp poses, guided by equivaraiance and residual actions obtained from the plan. These policies transfer zero-shot to the real world and achieve 80% success rates. For systematic evaluation, we propose a benchmark suite of multi-part assemblies resembling industrial and daily objects across diverse categories and geometries. By integrating efficient global planning and robust local control, we demonstrate the first system to achieve complete and generalizable real-world multi-part assembly without domain knowledge or human demonstrations.} }
Endnote
%0 Conference Paper %T Fabrica: Dual-Arm Assembly of General Multi-Part Objects via Integrated Planning and Learning %A Yunsheng Tian %A Joshua Jacob %A Yijiang Huang %A Jialiang Zhao %A Edward Li Gu %A Pingchuan Ma %A Annan Zhang %A Farhad Javid %A Branden Romero %A Sachin Chitta %A Shinjiro Sueda %A Hui Li %A Wojciech Matusik %B Proceedings of The 9th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2025 %E Joseph Lim %E Shuran Song %E Hae-Won Park %F pmlr-v305-tian25a %I PMLR %P 406--419 %U https://proceedings.mlr.press/v305/tian25a.html %V 305 %X Multi-part assembly poses significant challenges for robotic systems to execute long-horizon, contact-rich manipulation with generalization across complex geometries. We present a dual-arm robotic system capable of end-to-end planning and control for autonomous assembly of general multi-part objects. For planning over long horizons, we develop hierarchies of precedence, sequence, grasp, and motion planning with automated fixture generation, enabling general multi-step assembly on any dual-arm robots. The planner is made efficient through a parallelizable design and is optimized for downstream control stability. For contact-rich assembly steps, we propose a lightweight reinforcement learning framework that trains generalist policies across object geometries, assembly directions, and grasp poses, guided by equivaraiance and residual actions obtained from the plan. These policies transfer zero-shot to the real world and achieve 80% success rates. For systematic evaluation, we propose a benchmark suite of multi-part assemblies resembling industrial and daily objects across diverse categories and geometries. By integrating efficient global planning and robust local control, we demonstrate the first system to achieve complete and generalizable real-world multi-part assembly without domain knowledge or human demonstrations.
APA
Tian, Y., Jacob, J., Huang, Y., Zhao, J., Gu, E.L., Ma, P., Zhang, A., Javid, F., Romero, B., Chitta, S., Sueda, S., Li, H. & Matusik, W.. (2025). Fabrica: Dual-Arm Assembly of General Multi-Part Objects via Integrated Planning and Learning. Proceedings of The 9th Conference on Robot Learning, in Proceedings of Machine Learning Research 305:406-419 Available from https://proceedings.mlr.press/v305/tian25a.html.

Related Material