Sim-Suction: Learning a Suction Grasp Policy for Cluttered Environments Using a Synthetic Benchmark
Juncheng Li, David J. Cappelleri
Abstract
This article presents <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Sim-Suction</i> , a robust object-aware suction grasp policy for mobile manipulation platforms with dynamic camera viewpoints, designed to pick up unknown objects from cluttered environments. Suction grasp policies typically employ data-driven approaches, necessitating large-scale, accurately-annotated suction grasp datasets. However, the generation of suction grasp datasets in cluttered environments remains underexplored, leaving uncertainties about the relationship between the object of interest and its surroundings. To address this, we propose a benchmark synthetic dataset, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Sim-Suction-Dataset</i> , comprising 500 cluttered environments with 3.2 million annotated suction grasp poses. The efficient <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Sim-Suction-Dataset</i> generation process provides novel insights by combining analytical models with dynamic physical simulations to create fast and accurate suction grasp pose annotations. We introduce <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Sim-Suction-Pointnet</i> to generate robust 6-D suction grasp poses by learning point-wise affordances from the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Sim-Suction-Dataset</i> , leveraging the synergy of zero-shot text-to-segmentation. Real-world experiments for picking up all objects demonstrate that <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Sim-Suction-Pointnet</i> achieves success rates of 96.76%, 94.23%, and 92.39% on cluttered level 1 objects (prismatic shape), cluttered level 2 objects (more complex geometry), and cluttered mixed objects, respectively.