Abstract
Bin-picking of small parcels and other textureless planar-faced objects is a common task at warehouses. A general color image–based vision-guided robot picking system requires feature extraction and goal image preparation of various objects. However, feature extraction for goal image matching is difficult for textureless objects. Further, prior preparation of huge numbers of goal images is impractical at a warehouse. In this paper, we propose a novel depth image–based vision-guided robot bin-picking system for textureless planar-faced objects. Our method uses a deep convolutional neural network (DCNN) model that is trained on 15,000 annotated depth images synthetically generated in a physics simulator to directly predict grasp points without object segmentation. Unlike previous studies that predicted grasp points for a robot suction hand with only one vacuum cup, our DCNN also predicts optimal grasp patterns for a hand with two vacuum cups (left cup on, right cup on, or both cups on). Further, we propose a surface feature descriptor to extract surface features (center position and normal) and refine the predicted grasp point position, removing the need for texture features for vision-guided robot control and sim-to-real modification for DCNN model training. Experimental results demonstrate the efficiency of our system, namely that a robot with 7 degrees of freedom can pick randomly posed textureless boxes in a cluttered environment with a 97.5% success rate at speeds exceeding 1000 pieces per hour.
Highlights
Demand for logistics workers in Japan is high and still rising, making automatic robot manipulation systems crucially important for overcoming labor shortages [1]
We propose a deep convolutional neural network (DCNN) trained on 15,000 annotated depth images synthetically generated in a physics simulator [39] to predict grasp points and their grasp patterns, namely, the suction hand vacuum-cup activation mode: left cup on (L), right cup on (R), or both cups on (B)
Our system achieved a 97.5% success rate at a speed exceeding 1000 Piece per Hour (PPH), as demonstrated in the video (Supplementary File 1) in Supplementary Material. Both textured and textureless boxes could be held by a suction hand
Summary
Demand for logistics workers in Japan is high and still rising, making automatic robot manipulation systems crucially important for overcoming labor shortages [1]. Unlike two-stage methods, a one-stage method [28,29,30,31,32,33,34,35,36,37], namely, one-shot grasp detection, directly regresses grasp points and their classes without object segmentation or pose estimation This method is preferable for object picking in a warehouse for two reasons. We propose a novel depth-image-based vision-guided robot bin-picking system for textureless planar-faced objects for robots with a two-vacuum-cup suction hand. We propose a deep convolutional neural network (DCNN) trained on 15,000 annotated depth images synthetically generated in a physics simulator [39] to predict grasp points (centers of graspable surfaces) and their grasp patterns, namely, the suction hand vacuum-cup activation mode: left cup on (L), right cup on (R), or both cups on (B). We incorporate surface-feature descriptors for DCNN prediction refinement and feature extraction, allowing the system to be free of sim-to-real refinement for DCNN predictions, as well as texture features and goal images for VS
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
