Résumé
The purpose of this study is to develop a new system to synchronously control the simulated robot and humanoid robot based on the skeletal tracking technology. This system is a human-robot interaction system. Human can control the robot through the skeletal tracking technology without placing any marker on the subject's body. We use the Microsoft Kinect to get the skeleton framework information of the subject and input the data into the program. Therefore, the experiment has no demand on color to distinguish the subject's motions, and the subject can just perform in the front of the Kinect without wearing any control device. After processing the data, the computer controls the simulated robot to regenerate the subject's motions synchronously. In this study, we can control the simulated robot to do three different exercises. To verify the validity of the control system, we did a series of experiments to control a humanoid robot NAO. The experimental results show that the synchronous control system can distinguish the movements of human upper limb efficiently, and the cost is lower than that of the previous studies. This system controls both robot model and humanoid robot NAO based on human motion to achieve the human-machine interaction efficiently and the success rate is more than 90%.