DAWN: Diffusion-based Alignment and unfastening Workflow Network
DAWN: Diffusion-based Alignment and unfastening Workflow Network
This paper presents DAWN, a hierarchical robotic framework for performing real-world unfastening tasks through multi-modal sensory integration. The framework is divided into two stages: alignment & insertion, and unfastening, with three key components—AlignNet, LatchNet, and UnfastNet—ensuring precise engagement, reliable stage transitions, and stable force-controlled removal. Experimental results on a Franka Emika robot show success rates comparable to human operators, with ablation studies highlighting the method's scalability and data efficiency.
DAWN introduces a hierarchical robotic framework for unfastening tasks using diffusion-based policies and multi-modal sensing, demonstrating success rates comparable to human operators in simulation and real-world experiments.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
ensuring precise screwdriver tip engagement; LatchNet, a multi-modal classifier reliably detecting successful stage transitions; and UnfastNet, which achieves stable, force-controlled screw removal. Experimental validation on a Franka Emika robot demonstrates that DAWN achieves alignment & insertion success rates comparable to trained human operators, robustly transitions between stages, and attains an impressive su
Open source S4
ccess rate in unfastening across various unseen scenarios. Ablation studies further elucidate critical input modalities and effective training strategies, highlighting DAWN's scalability, robustness, and data efficiency.
Open source S5
Source：arXiv Robotics — research abstracts · arxiv.org