Digital twin model construction of robot and multi-object under stacking environment for grasping planning

Abstract

Object grasping is the basic function for robots to complete various tasks. However, it is still a hot and difficult problem to grasp object in multi-object stacking environment. A digital twin model construction method for robot grasping under multi-object stacking environment is proposed based on digital twin technology to realize multi-object grasping planning and virtual-real interaction. The robot digital twin model is constructed based on the requirements of the twin model, which includes the robot information model and information perception method. Then, the logical relationship between the virtual models of multi-object stacked objects is established based on the interactive relationship of stacked objects. The twin model of robot grasping under the multi-object stacking scene is built to realize the monitoring and visualization of the key elements of the physical entity of the robot based on ROS and Unity3D, and the interaction between the virtual model and the physical entity is realized to monitor the progress of robot grasping of multi-target objects. What’s more, the interaction relationship between objects is detected at the same time with an accuracy of 80 %. The experimental results verify the validity and reliability of the proposed method for establishing the digital twin model of robot grasping and the logical relation of the virtual model of objects.