Reproduce “A3D: Adaptive Affordance Assembly with Dual-Arm Manipulation” 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
Jiaqi Liang, Yue Chen, Qize Yu, Yan Shen, Haipeng Zhang, Hao Dong, Ruihai Wu
The A3D framework enables robots to perform generalizable furniture assembly by learning adaptive affordances for dual-arm coordination, where one arm stabilizes while the other manipulates. It demonstrates superior performance over baselines in both simulation and real-world scenarios across various assembly tasks, adapting support strategies dynamically based on interaction feedback.
Key insights
- Initial ambiguous affordance predictions for support locations become significantly more concentrated and precise after dynamic adaptation based on physical interaction feedback.
- Dynamic refinement of support strategies using interaction context, providing "physical awareness," is essential for robust dual-arm manipulation and generalization to unseen object categories.
- Utilizing a Top-K set of candidate support actions improves robustness by allowing exploration of multiple high-quality options, mitigating issues like collisions or unstable grasps.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.