A Hybrid Sensing Approach to Mobile Robot Localization in Complex Indoor Environments

Abstract

This paper presents a hybrid sensing system for mobile robot localization in large-scale indoor environments. The system operates in two sensing modes, either omni-directional vision or laser scanning, according to the environmental characteristics. For a structured corridor environment, the vision information is adopted to track the robot pose with a predefined hybrid metric-topological map. For a semi-structured office room, the laser scanning mode is chosen to generate a sequence of relative pose transformations based on a scan matching algorithm. Kalman filters are deployed to smooth multiple scan matching results. The proposed hybrid sensing system can perform localization tasks on-the-fly, with the features of efficient map modelling and computational simplicity. Experimental results are provided to demonstrate the performance and effectiveness of the proposed techniques.