Using Synthetic Data and Deep Networks to Recognize Primitive Shapes for Object Grasping
This addresses robotic manipulation challenges by enabling grasp prediction for both general and specific tasks, though it is incremental in using synthetic data and existing deep networks.
The paper tackles the problem of robotic object grasping by decomposing objects into primitive shapes from monocular depth input, achieving a 94% success rate in task-free grasping and 76% in task-oriented grasping.
A segmentation-based architecture is proposed to decompose objects into multiple primitive shapes from monocular depth input for robotic manipulation. The backbone deep network is trained on synthetic data with 6 classes of primitive shapes generated by a simulation engine. Each primitive shape is designed with parametrized grasp families, permitting the pipeline to identify multiple grasp candidates per shape primitive region. The grasps are priority ordered via proposed ranking algorithm, with the first feasible one chosen for execution. On task-free grasping of individual objects, the method achieves a 94% success rate. On task-oriented grasping, it achieves a 76% success rate. Overall, the method supports the hypothesis that shape primitives can support task-free and task-relevant grasp prediction.