RelAfford6D: Relational 6D Affordance Graphs for Constraint-Driven Robotic Manipulation
For robotic manipulation of articulated objects, this work provides a physically grounded alternative to data-driven methods, addressing the bottleneck of kinematic constraint satisfaction in open-world settings.
RelAfford6D introduces a training-free framework that uses a Relational 6D Affordance Graph to map semantic instructions to precise SE(3) poses, enabling constraint-driven robotic manipulation. It achieves superior zero-shot success rates, cross-category generalization, and execution robustness over data-driven baselines in simulation and real-world environments.
Bridging abstract semantics and precise physical control remains a fundamental challenge in open-world robotic manipulation. While recent data-driven policies show promise, their reliance on isolated contact points or latent affordance embeddings lacks the rigorous kinematic constraints necessary for complex articulated objects.To overcome the limitation, we introduce RelAfford6D, a novel training-free framework centered on a Relational 6D Affordance Graph. Given a free-form instruction, our system deduces a semantic topology linking a primary interacting part to its physical anchor. By elevating these topological nodes into precise metric $SE(3)$ poses via vision foundation models, we analytically formulate downstream execution as a kinematic constraint satisfaction problem. The robot synthesizes continuous trajectories by tracking strictly defined physical manifolds (e.g., revolute or prismatic orbits). Coupled with a closed-loop tracking mechanism for dynamic replanning against disturbances, our physically grounded approach achieves superior zero-shot success rates, cross-category generalization and execution robustness in both simulation and the real world environments, outperforming existing data-driven baselines.