Search papers, labs, and topics across Lattice.
This paper addresses the challenge of motion feasibility prediction in cluttered environments by learning directly from raw RGB-D observations, which is crucial for robotics applications like task and motion planning. The authors introduce a large-scale benchmark consisting of 2.7 million grasp feasibility labels across 88 objects and 190 cluttered scenes, enabling a comprehensive evaluation of various classifier architectures. Their best-performing model, GRASPFC-PTX, a point-cloud transformer, achieves an impressive AUROC of 0.996 on novel objects while significantly outpacing traditional sampling-based motion planners in prediction speed.
Learning motion feasibility from raw RGB-D data can achieve near-perfect accuracy while drastically reducing computational costs compared to traditional methods.
Motion feasibility prediction plays a central role in robotics, particularly in task and motion planning and manipulation. A major bottleneck for this problem in cluttered environments is that infeasible planning attempts by Sampling-based motion planners (SBMPs) can incur substantial computational cost. Also existing approaches for infeasibility certification are limited to low-dimensional configuration spaces and often assume simplified geometric environments represented by primitive objects with known parameters. We study the complementary problem of learning motion feasibility prediction directly from raw RGB-D observations for a 7-DOF manipulator operating in realistic cluttered scenes. We introduce the first large-scale benchmark for this setting, comprising 2.7M grasp feasibility labels over 88 scanned objects and 190 cluttered tabletop scenes. We benchmark three representative classifier families spanning MLP- based, volumetric-CNN, and point-cloud-based Transformer architectures under matched training conditions. Our best model, GRASPFC-PTX (a point-cloud transformer), achieves an AUROC of 0.996 on Novel objects while providing predictions significantly faster than SBMPs.
