Based on wireless local area networks (WLAN), this paper presents the navigation of a robotic wheelchair. We propose a radio based robot localization with a particle filter by using received signal strength from WLAN. An intelligent strategy is proposed to select the informative WLAN signals and reduce the algorithm complexity. Furthermore, we introduce a graphic feature node extraction to explore the topology and linkages of a given graphic map. Based on the radio positioning and graphic map features, path planning can be provided for the task environment to accomplish the robot navigation with a fuzzy based behavior decision. Thus, the robotic wheelchair could response to the real dynamical environment. The experimental results conducted on real WLAN measurements indicate that the proposed approach provides the accurate localization and auto navigation services.