大学院生のPham Ngoc Hungさんが、2017年度ARIS2017においてベストプレゼンテーション賞を受賞しました。
【受賞者】Pham Ngoc Hungさん（博士課程 機能制御システム専攻 ）
【指導教員】吉見 卓 教授（電気工学科）
Programming Everyday Task Using Primitive Skills and Generative Model of Movement Demonstrated by Human
The purpose of this research to propose an automated programming method of everyday manipulation task for robot.
In this research, I proposed to use a generative model (Dynamic Movement Primitives) of hand movement which is recorded from human demonstration and build a collection of primitive skills for the robot. The task can be automatically programmed from the sequence of primitive skills with adaptation to new changes such as the change of object’s location. In the experiment, a robot arm with a parallel gripper executed successfully the task ‘dispensing water’ from a water thermos pot by this programming method.
I expect to use this research result in developing intelligent service robots at home where the robots can accomplish everyday manipulation tasks. This research can be improved and developed so that the robot learn and execute automatically manipulation task from instructions and human demonstration.