Camera Calibration using Mobile Robot in Intelligent Space

Abstract

In this paper, we describe an automated camera calibration method by using mobile robots. In the recent years, the research field on smart environments, which is a space with many embedded and networked sensors, has been expanding. Such a system should have flexibility and scalability so that we can easily add, remove or replace devices depending on the size of the space, application, technological advances, and so on. One of the major problems when setting up the devices is a calibration of the sensors. In order to reduce this, we utilize color markers installed on a mobile robot to calibrate distributed vision sensors. The mobility of mobile robots allows us to cover wide areas of the environment without placing many landmarks in the exact positions beforehand. The position and orientation of the mobile robot are estimated from wheel encoder and an ultrasonic 3D positioning system measurement data using Extended Kalman Filter (EKF). Automated camera calibration is performed based on the positions of the color markers in world coordinate system and their corresponding points in image coordinate system. Experimental result shows the validity of this method.