A Vision-Based Intelligent Environment for Mobile Robot Localization and Navigation

An intelligent environment employing multiple vision sensors labeled with identification features is proposed. The union of the sensing ranges of these vision sensors covers a large unknown workspace where a mobile robot can proceed to explore. The goal of this research is to accomplish mobile robot localization and navigation based on such a calibration-free sensor network. Specifically, stationary CCD cameras mounted in the intelligent environment are to be located by a CCD camera onboard the mobile robot in terms of identification features. The position and sensing range of each newly located stationary CCD camera can be on-line determined based on the robot features observed by current and newly located stationary CCD cameras. The proposed system only requires a PC, an onboard CCD camera, and stationary CCD cameras equipped with wireless internet modules. Therefore, the proposed seemingly novel approach appears to be low-cost, effective, efficient, and calibration-free.