arXiv Robotics — research abstracts· Lizhi Yang, Yiling Hou, Yao Tang, Junheng Li, Daniel Weng, Blake Werner, Aaron D. Ames·· 22 hours agoEditorial score62
CSF: Contextual Safety Filtering for Motion Generators
CSF: Contextual Safety Filtering for Motion Generators
Summary
This research proposes CSF, a contextual safety filtering method for motion generators that integrates natural-language safety rules with reference trajectories to ensure safe motion execution. The approach successfully prevents unsafe motions in real-world scenarios, including human and object interactions, with up to 90% reduction in danger events.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
Source：arXiv Robotics — research abstracts · arxiv.org