BLaFA: Bridging Language to 3D Functional Affordances for Dexterous Robotic Grasping in Tool Use
BLaFA: Bridging Language to 3D Functional Affordances for Dexterous Robotic Grasping in Tool Use
This research paper presents BLaFA, a structured framework designed to enable dexterous robotic grasping in tool-use scenarios by bridging natural language instructions with 3D functional contact structures. The system utilizes a Knowledge-guided Language Parsing module to convert instructions into manipulation constraints, employs 3D Gaussian Splatting for hierarchical semantic representation and a Triangular Functional Point Localization module to identify contact keypoints, and finally uses a 3D Keypoint Grasp Matrix Transformation to generate executable wrist poses and joint commands. Evaluations across 100 open-vocabulary language instructions, 14 task-tool combinations, and real-world experiments demonstrate the framework's ability to translate language into task-aligned grasps for diverse tools. The source code will be made publicly available.
The paper introduces BLaFA, a framework that maps natural language instructions to 3D functional contact structures for robotic tool use, evaluated on 100 open-vocabulary instructions and 14 task-tool combinations with 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.
Reported numbers
instructions
100
Reported trials
View original evidence
ints to executable wrist poses and finger-level joint and force commands. Evaluations on 100 open-vocabulary language instructions, 14 task-tool combinations covering four functional intents, and real-world robotic experiments in representative scenes demonstrate that BLaFA can effectively translate natural-language instructions into task-aligned 3D functional contact structures and corresponding dexterous grasps, th
Open source S5task-tool combinations
14
Reported trials
View original evidence
ints to executable wrist poses and finger-level joint and force commands. Evaluations on 100 open-vocabulary language instructions, 14 task-tool combinations covering four functional intents, and real-world robotic experiments in representative scenes demonstrate that BLaFA can effectively translate natural-language instructions into task-aligned 3D functional contact structures and corresponding dexterous grasps, th
Open source S5functional intents
4
Reported trials
View original evidence
ints to executable wrist poses and finger-level joint and force commands. Evaluations on 100 open-vocabulary language instructions, 14 task-tool combinations covering four functional intents, and real-world robotic experiments in representative scenes demonstrate that BLaFA can effectively translate natural-language instructions into task-aligned 3D functional contact structures and corresponding dexterous grasps, th
Open source S5
- code: Reported pending Open source S6
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
ints to executable wrist poses and finger-level joint and force commands. Evaluations on 100 open-vocabulary language instructions, 14 task-tool combinations covering four functional intents, and real-world robotic experiments in representative scenes demonstrate that BLaFA can effectively translate natural-language instructions into task-aligned 3D functional contact structures and corresponding dexterous grasps, th
Open source S5
ereby enabling language-driven robotic tool use across diverse tasks. The source code will be made publicly available. Code will be publicly available at https://github.com/PopeyePxx/BLaFA.
Open source S6
Source：arXiv Robotics — research abstracts · arxiv.org