主催: 一般社団法人 日本機械学会
会議名: ロボティクス・メカトロニクス 講演会2017
開催日: 2017/05/10 - 2017/05/13
We research the best method to handover objects between a robot and humans for the daily necessities. In a handover motion, humans consider n the naturally and nature of the object to check the appropriate grasping and handover motion, so has to do the handover robotic system we have developed uses a Leap Motion sensing device to recognize the position and the posture of the human hands, and a gripping dimension method to classify the object gripping posture pattern. Our robotic system determines the object desired by human from the human motion and hand posture, and generate appropriate action with optimal speed positon, orientation, posture and timing.
In this research, we measured human-to-human handover, classified posture by object shape information, and created a database of grasping posture patterns of giver and receiver. Based on these results, we developed the handover robot system that human feel more natural to interact with.