In this paper a new approach to spatial knowledge management for a mobile robot is proposed. The framework merges two types of maps, grid and feature based, and two uncertainty representations, probabilistic and fuzzy. It is implemented by a set of logical sensors, that extract features from range sensors data, filter out artifacts, and estimate the robot displacement. Indoor simultaneous localization and mapping is considered as the target application
Published in:
Multisensor Fusion and Integration for Intelligent Systems, 2006 IEEE International Conference on
Date of Conference: Sept. 2006