Reproduce “BagIt! An Adaptive Dual-Arm Manipulation of Fabric Bags for Object Bagging” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Peng Zhou, Jiaming Qi, Hongmin Wu, Chen Wang, Yizhou Chen, Zeqing Zhang
Researchers from The University of Hong Kong, The Great Bay University, Northeast Forestry University, and Guangdong Academy of Sciences developed "BagIt!", an adaptive dual-arm robotic system for bagging diverse objects with fabric bags. The system employs a Structure-of-Interest strategy focusing on the bag's opening rim and achieves manipulation success rates up to 8/8 with an average SOI-to-object center error of 2.10 cm.
Key insights
- Focusing robotic manipulation on a simplified "Structure-of-Interest" (SOI), such as a bag's opening rim, can effectively manage the inherent complexity of deformable object interaction.
- Combining real-time visual perception with online-adapted models, like Broyden's approximation for the Deformation Jacobian Matrix within an MPC framework, enables robust control without explicit material property knowledge.
- A multi-stage planning and control approach, separating pre-bagging and bagging phases with constraint-aware trajectory generation, significantly enhances reliability and success rates for complex deformable object tasks.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.