Rearrangement Planning for General Part Assembly
This addresses the limitation of robotic assembly being restricted to single targets or categories, enabling more flexible and general autonomous assembly systems.
The paper tackles the problem of determining precise part poses for novel target assemblies with unseen part shapes, a task called rearrangement planning, and presents GPAT, a transformer-based model that achieves accurate pose predictions with demonstrated generalization to diverse novel shapes in both 3D CAD models and real-world scans.
Most successes in autonomous robotic assembly have been restricted to single target or category. We propose to investigate general part assembly, the task of creating novel target assemblies with unseen part shapes. As a fundamental step to a general part assembly system, we tackle the task of determining the precise poses of the parts in the target assembly, which we we term ``rearrangement planning''. We present General Part Assembly Transformer (GPAT), a transformer-based model architecture that accurately predicts part poses by inferring how each part shape corresponds to the target shape. Our experiments on both 3D CAD models and real-world scans demonstrate GPAT's generalization abilities to novel and diverse target and part shapes.