Method for determining position and posture of lifting appliance of intelligent gantry crane

Abstract

The invention discloses a method for determining the position and posture of a lifting appliance of an intelligent gantry crane, which comprises the following steps of: constructing a gantry crane digital model and a camera module so as to obtain a descending view angle video sequence of the lifting appliance of the gantry crane and construct a gantry crane view data training set; performing corner column calibration on the gantry crane view data training set to obtain a calibrated corner column value; inputting the view data training set into a neural network, and performing corner column detection by using a loss function to obtain a corner column position value; segmenting the corner column position value to obtain a corner column pixel space position; the depth value of the lifting appliance is obtained through a three-point model, and the length value and the width value of the lifting appliance are estimated through scale estimation so as to adjust the pose of the lifting appliance; and comparing the pixel space position of the corner column and the depth value, the length value and the width value of the lifting appliance through the gantry crane view data training set so as to judge whether the four corner columns of the lifting appliance are aligned with the container or not. Through corner column calibration and neural network training, the positioning error is reduced, and manual visual inspection is not needed for lifting appliance adjustment.

Description

technical field [0001] The invention relates to the field of intelligent gantry cranes, in particular to a method for determining the position and posture of a spreader of an intelligent gantry crane. Background technique [0002] At present, the container area and container location of the gantry crane in the container yard are completely operated manually by the driver. Manual observation and judgment of the container area and container location are prone to errors in identification of the container area and container location, resulting in information errors in container stacking, which brings great harm to container management. trouble. At the same time, the position of the container stacked by the gantry crane depends entirely on the manual visual inspection of the truck driver. There is no accurate positioning information between the gantry crane spreader and the container in the yard. Adjust the position of the gantry crane, which reduces the loading and unloading sp...

AI Technical Summary

Problems solved by technology

[0003] The invention provides a method for determining the position and posture of the spreader of an intelligent gantry crane to overcome the technical problems of relying on manual visual inspection for the position of containers stacked by the gantry crane

Images