This communication addresses the issue of concurrent map building and localization (CML) for a mobile robot in an unknown environment. The proposed solution extends over previous contributions in that the environment must not be static, nor the landmarks be uniquely identifiable. To this aim we introduce a map model that includes not only the robot and landmark locations in a reference frame, but also a model for landmark quality assessment. Convergence of the map covariance is preserved in the new map model.

This work was supported by projects: 'Navegación autónoma de robots guiados por objetivos visuales' (070-720), 'Supervised learning of industrial scenes by means of an active vision equipped mobile robot.' (J-00063).

International Conference on Pattern Recognition (ICPR), 2002, Quebec (Canadá)

Peer Reviewed