Method for Estimation of Grasping Point based on Geometric Features using Vision Sensor

摘要

The method for deriving grip points based on geometric features using a vision sensor according to an embodiment of the present invention is a method for deriving grip points based on geometric features performed by a processing device using an image acquired by a vision sensor as an input image, the method comprising: Searching for an effective gripping area using the outer surface of the object detected in the input image and evaluating the gripping index based on the shape of the object; searching for an effective gripping area for an isolated area using a depth map and evaluating a gripping index based on the isolated area; deriving an optimal gripping point using the gripping index evaluated based on the object shape and the gripping index evaluated based on the isolated area; and converting the derived optimal gripping point into a robot coordinate system.