Tracking System

Currently, our laboratory room, which is about $5[m]$ in both width and depth, is used for the ISpace. The ISpace has a mobile robot as a human-following agent, 6 DINDs which can obtain the situation in the environment, and a projector and a screen which present suitable information to the human. Each module is connected through the network communication.

Four DINDs are used in order to recognize the mobile robot and to generate the control commands. The other two DINDs are used to recognize the position of the human. DINDs are placed as shown in Fig2(a). Fig.2(b) is a picture of the actual ISpace. The placement of the two DINDs for human recognition is optimized to expand the viewable area of the cameras[12] so that the head and hands of the human can be recognized over a wide area. On the other hand, the placement of DINDs for the mobile robot has to be decided by trial and error. It is desirable that the DINDs for the mobile robot recognize the whole of the area covered by two DINDs for human recognition in order to achieve the human-following system and reliable mobile robot control. Thus, four DINDs are placed so that the area for human recognition is completely covered. Human walking information is extracted by background subtraction and by detecting the skin color of a face and hands on captured images. A 3D position is reconstructed by stereo vision using two cameras.

In the coordinate system of the ISpace, the x axis is parallel to the screen, the y axis is perpendicular to it, and the z axis follows the right-hand system. Each DIND measures 3D positions, which are based on this coordinate system.