AffordGrasp: In-Context Affordance Reasoning for Open-Vocabulary Task-Oriented Grasping in Clutter

Inferring the affordance of an object and grasping it in a task-oriented manner is crucial for robots to successfully complete manipulation tasks. Affordance indicates where and how to grasp an object by taking its functionality into account, serving as the foundation for effective task-oriented grasping. However, current task-oriented methods often depend on extensive training data that is confined to specific tasks and objects, making it difficult to generalize to novel objects and complex scenes. In this paper, we introduce AffordGrasp, a novel open-vocabulary grasping framework that leverages the reasoning capabilities of vision-language models (VLMs) for in-context affordance reasoning. Unlike existing methods that rely on explicit task and object specifications, our approach infers tasks directly from implicit user instructions, enabling more intuitive and seamless human-robot interaction in everyday scenarios. Building on the reasoning outcomes, our framework identifies task-relevant objects and grounds their part-level affordances using a visual grounding module. This allows us to generate task-oriented grasp poses precisely within the affordance regions of the object, ensuring both functional and context-aware robotic manipulation. Extensive experiments demonstrate that AffordGrasp achieves state-of-the-art performance in both simulation and real-world scenarios, highlighting the effectiveness of our method. We believe our approach advances robotic manipulation techniques and contributes to the broader field of embodied AI.

Overall Framework of AffordGrasp. The framework processes user instructions and RGB-D scene observations to achieve open-vocabulary task-oriented grasping in clutter. We leverage GPT-4o for in-context affordance reasoning, decomposing the process into three steps: (1) Extracting the task goal and functional requirements from implicit user instructions (e.g., "I want to scoop something"). (2) Identifying the most task-relevant object in the RGB image (e.g., a wooden spoon). (3) Decomposing the object into functional parts and selecting the optimal graspable part (e.g., the handle) based on its affordances. Based on the reasoning results, a visual affordance grounding module grounds the inferred object and part affordances into pixel-level masks. With the affordance mask and RGB-D images, we employ AnyGrasp to generate task-oriented 6D grasp poses on the target part.