A soft bag sorting and grabbing point determination method based on a three-dimensional point cloud

Abstract

The application provides a soft bag sorting and grabbing point determination method based on three-dimensional point cloud, and relates to the technical field of sorting robots. The method comprises the following steps: obtaining three-dimensional point cloud data of a soft bag stacking area through a depth camera, fitting a reference plane based on a soft bag bearing surface, establishing a height field representing the undulating state of the soft bag surface based on the reference plane, performing effective point screening, boundary rejection and smoothing and noise reduction processing on the height field to obtain a candidate area, and then performing wrinkle feature analysis or local peak prominence degree analysis on the candidate area to determine the optimal grabbing point coordinates and output. The method does not depend on color, texture or printed pattern information, can predict the optimal grabbing point through the geometric characteristics of the soft bag itself, improves the success rate, stability and sorting efficiency of automatic grabbing of the soft bag, and is suitable for automatic sorting and grabbing of flexible bag-shaped or film-shaped targets such as express soft packaging bags, woven bags and waste plastic bags.

Images