The Generalization of Robot Skills Based on Dynamic Movement Primitives ?

摘要

Learning from demonstration can transfer the operation skills of humans to robots and reach the purpose of fast and efficient programming to achieve flexible operation. It has the advantages of high programming efficiency, easy optimization and non-specialists can operate. And with the development of machine vision technology, the technology of object recognition based on machine vision is becoming more and more mature. This paper creatively combines the learning from demonstration based on dynamic movement primitives model with the object recognition technology based on visual information. Therefore, the robot can recognize the object automatically and generalize the skill. The feasibility and practicability of this method are verified by the experiment of robot finishing the desktop.