B. Kuipers, Y.T. Byun, A robust qualitative method for robot spatial learning, in: Proc. AAAI-88, St. Paul, MN, 1988, pp. 774--779.  Home/Search   Document Details and Download   Summary   Related Articles   Check

.... Intelligence Laboratory of the University of Texas at Austin has acquired two mobile robots for research purposes, specifically for research in mobile robot exploration, mapping and navigation using the Spatial Semantic Hierarchy representation of spatial knowledge [Kuipers and Levitt, 1988] [Kuipers and Byun, 1988] [Lee, 1994] One of the mobile robot is developed locally which we named Rover, while the other is a commercial mobile robot from Real World Interface TM (RWI) which we named Spot. This guide supplements the guide [Lee, 1995] in describing a high level programming package for programming the ....

Benjamin J. Kuipers and Yung-Tai Byun, "A robust, qualitative method for robot spatial learning," in Proc. AAAI-88, St.Paul, Minneapolis, 1988, pp.774-779.

....effectiveness in controlling physical mobile robots in real unstructured office environment is completed and reported here. Previous work developed the SSH only in outline and tested it only in simulated environments with very simple models of sensorimotor error using a simulated robot named NX [Kuipers and Byun, 1987, 1988, 1991] Physical implementations of aspects of the SSH approach have been built by other researchers but they do not provide adequate demonstration of its strengths, nor adequate analysis of its conditions of applicability. The underlying goal of this research is to demonstrate how symbolic ....

....is then used for route planning or similar problem solving purposes. Work by Chatila and Laumond [1985] exemplifies such an approach. Such an approach is often fragile in the face of sensory errors and effector inaccuracy. Its general drawbacks have been discussed in the works of Brooks [1985] Kuipers and Byun [1987, 1988, 1991] and Levitt et al. 1987] Byun [1990] and Thrun and Bucken [1996] To overcome the limitations of the traditional mapping method, Kuipers pioneered an approach based on a computational model of the human cognitive map that emphasized the distinction between control, procedural 2 , ....

[Article contains additional citation context not shown here]

Benjamin J. Kuipers & Yung-Tai Byun, "A robust, qualitative method for robot spatial learning," in Proc. 7th National Conference in Artificial Intelligence (AAAI-88), St.Paul, Minneapolis, 1988, pp. 774-- 779.

....we assume that it is available in terms of: Descriptions of typical sensor patterns present in the environment. Selection of patterns corresponding to distinctive or significant categories of objects and places enables recognition of landmarks through a sensor based classification algorithm [5] [10] Examples of significant categories are computers and telephones in office environments, buildings and monuments in urban environments. Characterization of semantically significant clusters of objects or places, obtained through sensor based recognition of passageways between different ....

B.J. Kuipers and Y.T. Byun, "A robust, qualitative method for robot spatial learning", in Proc. AAAI88, Saint Paul, Minnesota, vol. 2, pp. 774-779, 1988.

....the grid based approach and the topological approach. The first one [3] uses an accurate description of the environment that is based on a grid tessellation where each grid cell denotes a small and precisely pre determined region of the total space. On the other hand, topological approaches [10, 12, 8] build maps as graphs where nodes depict distinctive places in the environment and vertices correspond to paths between places. A recent approach that combines the above two approaches was proposed in [16] where a topological map was built on top of a grid based one. A relatively new area in the ....

Kuipers B.J., Byun Y.T.: A robust qualitative method for robot spatial learning. In: Proc. AAAI, (1988) 774--779.

....Furthermore, most of these systems evolve into highly specialized architectures with limited applicability [8] We extend the reactive approach with general map construction and handling capabilities for indoor navigation. Qualitative representations of sensor behavior have been used in [6] and [5] for spatial learning. Both approaches use these representations for landmark recognition, although only [5] was implemented on a real robot. We use qualitative representations for both planning and position control. Furthermore we use qualitative simulation for off line map construction. ....

B. J. Kuipers, Y-T. Byun, A Robust Qualitative Method for Robot Spatial Learning, Proceedings of AAAI-88, 1988.

....Furthermore, most of these systems evolve into highly specialized architectures with limited applicability [9] We extend the reactive approach with general map construction and handling capabilities for indoor navigation. Qualitative representations of sensor behavior have been used in [6] and [5] for spatial learning. Both approaches use these representations for landmark recognition. We use qualitative representations for both recognition and position control. 8] describes a planner that uses localized navigation templates for outdoor navigation. Our topological planner uses a ....

B. J. Kuipers, Y-T. Byun, A Robust Qualitative Method for Robot Spatial Learning, Proceedings of AAAI-88, 1988.

....of the respective Voronoi cells, are connected into a mesh by using the Delaunay triangulation. Then, we show how the basic robot tasks can be integrated on the resulted graph. 1 Introduction Exploring the free space and building a good map representation is a challenging mobile robot problem [1, 2, 3]. In this paper we propose a method by which the robot explores and learns its environment without any prior information about it. The notion of learning implies a way of modeling the free space so that subsequent tasks, like path planning or obstacle avoidance, can easily be carried out based ....

B. J. Kuipers and Y.-T. Byun, A robust qualitative method for robot spatial learning, Proc. AAAI (1988) 774--779.

....often required to describe the navigation task itself, e.g. to specify the destination or some required intermediate points. By contrast, some other systems also attempt to learn from observations the significant features of various locations, which may correspond to the set of landmarks; cf. [19], 23] 1 and others. Notice that our landmark selection approach (embodied in LearnSF) is complementary to those landmark acquiring systems: while those other systems acquire a set of landmarks that seem useful in some situations and so may be useful in general, our LearnSF algorithm can then ....

B. J. Kuipers and Y.-T. Byun, "A robust, qualitative method for robot spatial learning," Proceedings of the Seventh National Conference on Artificial Intelligence, pp. 774--779, Morgan Kaufmann, 1988.

....systems [1] Our autonomous wheelchair robot and its underlying qualitative navigation schema are described in more detail in [7] and [8] The performance of the system is therein evaluated and discussed upon real experiments. Qualitative representations of sensor behavior have been used in [4] and [5] for spatial learning. Both approaches use these representations for landmark recognition, although only [5] was implemented on a real robot. We use qualitative representations for both planning and position control. Furthermore we use qualitative simulation for off line map construction. ....

B. J. Kuipers and Y.-T. Byun. A robust qualitative method for robot spatial learning. In Proceedings of AAAI-88, 1988.

No context found.

B. Kuipers, Y.T. Byun, A robust qualitative method for robot spatial learning, in: Proc. AAAI-88, St. Paul, MN, 1988, pp. 774--779.

No context found.

B. J. Kuipers and Y.-T. Byun, `A robust, qualitative method for robot spatial learning', in AAAI 88. Seventh National Conference on Artificial Intelligence, pp. 774--779, (1988).

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC