Simultaneous localization and map building: a global topological model with local metric maps

Abstract

In this paper an approach combining the metric and topological paradigm for simultaneous localization and map building is presented The main idea is to connect local metric maps by means of a global topological map. This allows a compact environment model which does not require global metric consistency and permits both precision and robust- ness. The method uses a 360 degree laser scanner in order to extract comers and openings for the topological approach and lines for the metric localization. The approach has been tested in a 30x 25 m portion of the institute building with the fully autonomous robot Donald Duck An experiment consists of a complete exploration and a set of test missions. Three experiments have been performed for a total of 15 test missions, which have been randomly defined and completed with a success ratio of 87%.

A new technique in mobile robot simultaneous localization and mapping

“…Researchers have addressed this problem for well-structured (indoor) environments and have obtained important results (Anousaki et al 1999, Castellanos et al 1998, Kruse et al 1996, Leonard et al 1991, Thrun et al 2000, Tomatis et al 2001). These algorithms have been implemented for several different sensing methods, such as stereo camera vision systems (Castellanos et al 1998, Se et al 2002, laser range sensors (Tomatis et al 2001), and ultrasonic sensors (Anousaki et al 1999, Leonard et al 1991. Sensor movement/placement is usually done sequentially (raster scan type approach), by following topological graphs, or using a variety of greedy algorithms that explore regions only on the extreme edges of the known environment (Anousaki et al 1999, Leonard et al 1991.…”

“…Several localization schemes have been implemented, including topological methods such as generalized Voronoi graphs and global topological maps (Tomatis et al 2001), extended Kalman filters (Anousaki et al 1999, Leonard et al 1991, and robust averages. Although novel natural landmark selection methods have been proposed (Simhon et al 1998), most SLAM architectures rely on identifying landmarks as corners or edges in the environment (Anousaki et al 1999, Castellanos et al 1998, Leonard et al 1991.…”

Self-localization in Large-Scale Environments for the Bremen Autonomous Wheelchair

“…The resulting self-localization methods also work probabilistically on the basis of the odometry and a local model of the environment perceived with the sensors. A very recent approach by Tomatis et al combines map-building and self-localization [27]. They employ a 360 • laser range finder and extract features such as corners and openings which are used to navigate in a global topological map.…”

“…A number of prominent self-localization algorithms use the Markov localization approach, some of them with toplogical representations of the environment [23,15,27], others with metric maps [2,7,25]. In the robotics community, it is referred to as "Markov localization" if the algorithm somehow exploits the socalled Markov assumption [22].…”

Hybrid simultaneous localization and map building: closing the loop with multi-hypotheses tracking

“…The effectiveness of this method for localization has already been shown in [19]. In [20] the extension to simultaneous localization and map building is presented. In this paper it is shown how loops can be closed within the same framework.…”