Portal de Periódicos da CAPES

Efficiently Learning Single-Arm Fling Motions to Smooth Garments

2023; Springer International Publishing; Linguagem: Inglês

10.1007/978-3-031-25555-7_4

2511-1264

Lawrence Yunliang Chen, Huang Huang, Ellen Novoseller, Daniel Seita, Jeffrey Ichnowski, Michael Laskey, Richard K. Cheng, Thomas Kollar, Ken Goldberg,

Computer Graphics and Visualization Techniques

Recent work has shown that 2-arm “fling” motions can be effective for garment smoothing. We consider single-arm fling motions. Unlike 2-arm fling motions, which require little robot trajectory parameter tuning, single-arm fling motions are very sensitive to trajectory parameters. We consider a single 6-DOF robot arm that learns fling trajectories to achieve high garment coverage. Given a garment grasp point, the robot explores different parameterized fling trajectories in physical experiments. To improve learning efficiency, we propose a coarse-to-fine learning method that first uses a multi-armed bandit (MAB) framework to efficiently find a candidate fling action, which it then refines via a continuous optimization method. Further, we propose novel training and execution-time stopping criteria based on fling outcome uncertainty; the training-time stopping criterion increases data efficiency while the execution-time stopping criteria leverage repeated fling actions to increase performance. Compared to baselines, the proposed method significantly accelerates learning. Moreover, with prior experience on similar garments collected through self-supervision, the MAB learning time for a new garment is reduced by up to 87%. We evaluate on 36 real garments: towels, T-shirts, long-sleeve shirts, dresses, sweat pants, and jeans. Results suggest that using prior experience, a robot requires under 30 min to learn a fling action for a novel garment that achieves 60–94% coverage. Supplementary material can be found at https://sites.google.com/view/single-arm-fling .