Bimanual Shelf Picking Planner Based on Collapse Prediction

In logistics warehouse, since many objects are randomly stacked on shelves, it becomes difficult for a robot to safely extract one of the objects without other objects falling from the shelf. In previous works, a robot needed to extract the target object after rearranging the neighboring objects. In contrast, humans extract an object from a shelf while supporting other neighboring objects. In this paper, we propose a bimanual manipulation planner based on collapse prediction trained with data generated from a physics simulator, which can safely extract a single object while supporting the other objects. We confirmed that the proposed method achieves more than 80% success rate for safe extraction by real-world experiments using a dual-arm manipulator.

[1]  Nanning Zheng,et al.  Visual Manipulation Relationship Network for Autonomous Robotics , 2018, 2018 IEEE-RAS 18th International Conference on Humanoid Robots (Humanoids).

[2]  Li Fei-Fei,et al.  ImageNet: A large-scale hierarchical image database , 2009, CVPR.

[3]  Erik Schaffernicht,et al.  Support relation analysis and decision making for safe robotic manipulation tasks , 2015, Robotics Auton. Syst..

[4]  Changjoo Nam,et al.  Fast and resilient manipulation planning for target retrieval in clutter , 2020, 2020 IEEE International Conference on Robotics and Automation (ICRA).

[5]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  David Hsu,et al.  Act to See and See to Act: POMDP planning for objects search in clutter , 2016, 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[7]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[8]  Tamim Asfour,et al.  Active Vision for Extraction of Physically Plausible Support Relations , 2019, 2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids).

[9]  Bin Fang,et al.  Deep Reinforcement Learning for Robotic Pushing and Picking in Cluttered Environment , 2019, 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[10]  Xinyu Liu,et al.  Dex-Net 2.0: Deep Learning to Plan Robust Grasps with Synthetic Point Clouds and Analytic Grasp Metrics , 2017, Robotics: Science and Systems.

[11]  Hui Cheng,et al.  Learning Affordance Space in Physical World for Vision-based Robotic Object Manipulation , 2020, 2020 IEEE International Conference on Robotics and Automation (ICRA).

[12]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[13]  Amir Degani,et al.  Decision-making algorithms for safe robotic disassembling of randomly piled objects , 2017, Adv. Robotics.

[14]  T. Rabbani,et al.  SEGMENTATION OF POINT CLOUDS USING SMOOTHNESS CONSTRAINT , 2006 .

[15]  ChangHwan Kim,et al.  Efficient Obstacle Rearrangement for Object Manipulation Tasks in Cluttered Environments , 2019, 2019 International Conference on Robotics and Automation (ICRA).

[16]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[17]  Joni Pajarinen,et al.  Robotic manipulation in object composition space , 2014, 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[18]  Nanning Zheng,et al.  A Multi-task Convolutional Neural Network for Autonomous Robotic Grasping in Object Stacking Scenes , 2018, 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).