The usage of geometrical 3D-models enables humans to plan and design different aspects of indoor environments. This article describes the next-best-view planner for an autonomous 3D modelling robot using a 3D laser scanner. Besides the human user, the constructed 3D-model can also be used by other autonomous robots for navigation and localization tasks. The process of building a 3D-model without any geometical pre-knowledge leads to a planning scheme which uses the 3D data. This article copes with that problem by using and attributed 2D grid combined with an intelligent action planning scheme for indoor environments.
This paper introduces a new framework for incremental localization and mapping (3D-SLAM) that integrates submaps of surface features extracted from dense range images. Its local layer estimates unknown motion and feature associations between adjacent views. The global layer tries to ensure map consistency and to close loops, using multiple hypotheses offered by the local layer. The local algorithm crucially affects the map quality. An interpretation tree (IPT) is compared to Orthogonal Surface Assignment (OSA), a new algorithm tracing a building coordinate system inside man-made work spaces. During indoor experiments, performed with the rotating laser scanner RoSi, OSA proved much more reliable than IPT. Another experiment, running a simple on-line surface classification, indicates that the maps are useful for mission planning.Dieser Beitrag stellt ein neues Schema zur gleichzeitigen Lokalisierung und Kartierung (3D-SLAM) vor, das verdichtete Karten von Flächenmerkmalen aus Tiefenbildfolgen Schritt haltend erzeugt. Die lokale Ebene schätzt räumliche Messwege und Merkmalzuordnungen zwischen benachbarten Ansichten; die globale Ebene versucht die Kartenkonsistenz zu gewährleisten und Kreise zu schließen. Der Einfluss des lokalen Algorithmus auf die Gesamtleistung wird hier anhand eines Suchbaumes (IPT) und eines neuen Verfahrens für Gebäudestrukturen, Orthogonale Flächenzuordnung (OSA), untersucht. Experimente mit dem rotierenden Laserscanner RoSi in Gebäuden zeigten für OSA eine deutlich höhere Zuverlässigkeit als für IPT. Die Karten eignen sich für die Missionsplanung, was anhand einer einfachen Objektklassifizierung gezeigt wird.
This paper covers the global aspects of a new SLAM framework introduced in [9]. The map is a graph of overlapping range views, linked by multiple uncertain hypotheses of motion and correspondence. They form the interface between local pose estimation and globally consistent mapping. In order to prune the hypotheses and reduce the brittleness of the local algorithms, we propose a novel cooperation between image-based and odometric motion estimates, and a geometric-probabilistic visibility model for oriented surface features which can also discern moving objects. Global loop closing works by exchanging hypotheses, priorized on a node ambiguity measure to bound the update complexity. The cycle error in frame space is used as a consistency criterion. The new concepts were tested and evaluated during several indoor exploration tours.Dieser Beitrag behandelt den globalen Teil eines neuen Schemas [9] für schritthaltendes 3-D-SLAM. Die Karte ist ein Graph überlappender Teilansichten, welche durch unsichere mehrfache Hypothesen für Lage und Korrespondenz elastisch gekoppelt sind. Letztere bilden die Schnittstelle zwischen lokaler und globaler Ebene. Um die Sprödigkeit der lokalen Lageschätzung zu reduzieren, werden bildbasierte und odometrische Schätzwerte auf eine neue Art verküpft; ferner wird ein geometrisch-probabilistisches Modell der Sichtbarkeit für Flächenmerkmale eingeführt. Global werden Kreise durch Austausch von Lagehypothesen geschlossen, priorisiert nach dem Grad ihrer Mehrdeutigkeit und mit dem Kreisfehler im Lageraum als Konsistenzkriterium. Die Konzepte wurden an realen Explorationsfahrten in Gebäuden erprobt und evaluiert.
Efficient navigation of mobile platforms in dynamic human-centered environments is still an open research topic. We have already proposed an architecture (MEPHISTO) for a navigation system that is able to fulfill the main requirements of efficient navigation: fast and reliable sensor processing, extensive global world modeling, and distributed path planning. Our architecture uses a distributed system of sensor processing, world modeling, and path planning units. In this arcticle, we present implemented methods in the context of data fusion algorithms for 3D world modeling and real-time path planning. We also show results of the prototypic application of the system at the museum ZKM (center for art and media) in Karlsruhe.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.