Comparisons of planar detection for service robot with RANSAC and region growing algorithm

Zhang, Zhijun; Huang, Yongqian; Zhang, Wenkang; Luo, Jiawei

doi:10.23919/chicc.2017.8029128

Cited by 2 publications

(1 citation statement)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first is to refine g opt prediction by the learning-based grasp planning loop to g opt , which the surface center g opt belongs to, at the robot's home position in the first iteration. The second is to provide the surface center and normal to the 2.5D VS. To calculate surface features, we use the region-growing algorithm for surface construction, because region growing is more robust than RANSAC in actual applications [44]. To determine the region expansion rule, an initial seeking point g and the normal of the surface to which g belongs must be calculated.…”

Section: Visual-guided Robot Controlmentioning

confidence: 99%

Depth Image–Based Deep Learning of Grasp Planning for Textureless Planar-Faced Objects in Vision-Guided Robotic Bin-Picking

Jiang

Ishihara

Sugiyama

et al. 2020

Sensors

View full text Add to dashboard Cite

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.

show abstract

Section: Visual-guided Robot Controlmentioning

confidence: 99%