C. Taylor and D. Kriegman, "Vision-based motion planning and exploration algorithms for mobile robots," IEEE Tran. on Robotics and Automation, vol. 14, pp. 417--427, 1996.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....robot pose has to be estimated from the same measurements used for the mapping (simultaneous localization and mapping) An additional challenge in unknown environments is the issue of visual coverage or better known as visual exploration. We emphasize the visual aspect of coverage (as in [10] [17]) as opposed to area coverage meant either as producing a roadmap [7] or sweeping of space, for example in the case of vacuum cleaners or landmine detection. Usually many of the general exploration algorithms produce a redundant visual coverage and are thus inefficient if visual coverage and ....

C. Taylor and D. Kriegman. Vision-based motion planning and exploration algorithms for mobile robots. IEEE Trans. On Robotics and Automation, 14(3):147--427, 1998.

....in part by the National Science Foundation under POWRE grant EIA 9806108, VTEPSCoR grant OSR 9350540, by Middlebury College, and by a grant to Middlebury College from the Howard Hughes Medical Institute. 1. 2 Related work The approach taken in this paper and by a number of other research groups [1, 12, 6] is to use artificial landmarks that can be easily and unobtrusively added to the environment. Techniques for mobile robot navigation based on landmarks include those that are primarily reactive [3] those planned within a geometric environment map enhanced with perceptual landmarks [5, 7] and ....

C. J. Taylor and D. J. Kriegman. Vision-based motion planning and exploration algorithms for mobile robots. In Algorithmic Foundations of Robotics (Proceedings WAFR'94), pages 69--83, 1995.

....accurate geometric picture of the obstacles in the immediate surroundings, using only sensor readings which each convey a limited amount of unreliable information, with less emphasis on the notion 9 that the map should be stable over time. At the opposite extreme are systems like those of [52], which attempt to retrieve stable data on the topology of the obstacles in the environment, but which does not attempt to recover metric data at all. The emphasis here is entirely on a map which can be used to stably guide navigation over time, and while the data recovered is indeed useful for ....

.... much (to be precise, so long as it stays within the same cell of the aspect graph defined by all the other landmarks) 38 The major problem, as described above, is the difficulty with identifying the landmarks. This difficulty can be worked around in various ways for instance, consider [52], whose landmarks, for a system to be discussed below which faces some of the same difficulties, are pieces of paper taped on the walls of his lab, with designs on them which are chosen specifically so that they can be easily segmented and classified. Likewise, if the robot is a planetary ....

[Article contains additional citation context not shown here]

C. J. Taylor and D. J. Kriegman. Vision-based motion planning and exploration algorithms for mobile robots. In Workshop on the Algorithmic Foundations of Robotics, 1994.

....the associated visibility region with each landmark and boundary leading to a landmark we obtain partitioning of the environment in terms of places 3 . Two landmarks are in the neighborhood of each other if their visibility regions overlap. The overlapping regions are referred to as gateways [TK94]. For the successful navigation between the places we have to make sure that the setpoints of the elementary servoing strategies are chosen in such a manner that they will bring the camera frame (or the mobile base frame for that matter) to a gateway region of two landmarks (see Figure 4 for ....

C. J. Taylor and D. J. Kriegman. Vision-based motion planning and exploration algorithms for mobile robots. In Proceedings of the Workshop on the Algorithmic Foundations of Robotics, 1994.

....industrial automation actions. A common way to reduce the complexity of the environment is to put up aids for perception for navigation purposes, for example artificial landmarks like reflex tape in the ceiling (Robertson, 1991) Stanford University Hospital p. 50) or large bar codes on the walls (Taylor and Kriegman, 1995) (Yale p. 53) 3 Locomotion The most basic capability of a mobile agent is of course its ability to move. In nature we can observe flying, gliding, dolphin or turtle like swimming, walking, jumping, crawling, etc. For machines wheeled locomotion has been constructed as well. The choice one makes ....

....valued recognition functions, functions that are invariant to viewpoint, are constructed for sets of features and used as landmarks (Yeh and Kriegman, 1995) A goal for mobile robotics is to let the robot explore and map the environment by itself. This group is doing research to make this possible (Taylor and Kriegman, 1995). The building of local maps around landmarks, one at a time, is done using odometry and ultra sonic sensors to follow the boundary of the obstacles in the environment around the landmark. In their current implementation, cameras are used to see the landmarks, which consist of large bar codes to ....

Taylor, C.J. and Kriegman, D.J. (1995). Vision-based motion planning and exploration algorithms for mobile robots. In Goldberg, Halperin, Latombe, and Wilson, editors, The algorithmic Foundation of Robotics, pp. 69--84, Boston. A.K. Peters.

....due to their rich description of a scene and predictable response to changes in viewing position. 1 Introduction Many research groups have attempted to build autonomous visually guided robots. These system can generally be divided into two categories, a top down approach based on world models [4, 5] and a bottom up approach based on layered responses (such as the subsumption architecture [14] The first of these is limited by its ability to deal with the complexities of the real world and is generally restricted to indoor or man made environments. It has been remarked that such approaches ....

C.Taylor and D.Kriegman. Vision-based motion planning and exploration algorithms for mobile robots. Tech.rep. Yale University, 1993.

....as the robot travels. The difficulty of determining exact position has recently led robotics researchers to look for navigation algorithms that do not need exact global position, but use weaker information that can be determined without cumulative error. In this connection, Taylor and Kriegman [8] introduced the idea of distance queries, based on the capabilities of their robot RJ. The general question of how much and what kinds of sensory information a robot needs for various tasks constitutes an exciting area of research; the main contribution of this paper is to explore the implications ....

....on distances. This is potentially useful in practice, as discussed below. 2. Related Results. Previous theoretical research on online robot navigation has progressed in at least two research communities: the online competitive analysis community [1, 2, 6] and the theoretical robotics community [3, 4, 5, 8]. The competitive analysis community has generally considered various restricted types of polygonal obstacles and has measured the performance of a navigation algorithm by the ratio of the distance traveled by a robot using that algorithm to the length of the shortest obstacle avoiding path from S ....

[Article contains additional citation context not shown here]

C. J. Taylor and D. J. Kriegman, Vision-based motion planning and exploration algorithms for mobile robots, in Proc. of the Workshop on Algorithmic Foundations of Robotics, February 1994, San Francisco.

....with respect to the agent. These expectations can be used to drive the PA layer s actions. A round of tag proceeds as follows. The planner uses its map of the environment to determine a search path such that every point in the environment will eventually be in Bruce s absolute field of view [30]. An example plan is shown in figure 9. The solid lines represent paths to travel, while the dashed lines represent points where the agent must turn its camera in the direction indicated by the arrow. The agent is shown as a shaded pentagon with the point indicating its direction of motion. The ....

Taylor, C.J.; and Kriegman, D.J. 1994. Vision-Based Motion Planning and Exploration Algorithms for Mobile Robots. Workshop on the Algorithmic Foundation of Robotics.

....sensing: knowing where to look can improve the speed and reliability of the mapping process. Graph based techniques smear out exactness somewhat by relating position regions to nodes or places in a connectivity graph, and rely on closed loop feedback to guide the robot from place to place. In (Taylor Kriegman 1994) the model of the environment is a graph constructed while exploring, based on visibility of barcode landmarks. Similarly, in (Kortenkamp et al. 1992) Kuipers Byun 1981) and (Engelson 1994) the robot learns a model of the environment by building a graph based on topological connectivity of ....

Taylor, C., and Kriegman, D. 1994. Vision-Based Motion Planning and ExplorationAlgorithms for Mobile Robots. Workshop on the Algorithmic Foundations of Robotics.

....to explore a room with polygonal obstacles and the goal is to minimize the total distance traveled. Recently, Berman et al. BBF 96] investigated the case where the obstacles are oriented rectangles. Other work relating to exploring geometric domain are be found in [LS86, LS87, EBEY92, LT94, TK94, AWZ96] Betke et al. BRS95] and Awerbuch et al. ABRS95] investigated the problem with the additional constraints that the robot is required to return to its starting point for refueling periodically. Bender and Slonim [BS94] illustrated how two robots can collaborate in exploring directed ....

C. J. Taylor and D. J. Kriegman. Vision-based motion planning and exploration algorithms for mobile robots. In Proc. of the Workshop on Algorithmic Foundation of Robotics, 1994. This article was processed using the L a T E X macro package with LLNCS style

No context found.

C.J. Taylor and D. Kriegman. Vision-based motion planning and exploration algorithms for mobile robots. IEEE Trans. On Robotics and Automation, 14(3):147--427, 1998.

No context found.

C. Taylor and D. Kriegman, "Vision-based motion planning and exploration algorithms for mobile robots," IEEE Tran. on Robotics and Automation, vol. 14, pp. 417--427, 1996.

No context found.

C. J. Taylor and D. J. Kriegman, "Vision-based motion planning and exploration algorithms for mobile robots," IEEE Transactions on Robotics and Automation 14, pp. 417--426, June 1998.

No context found.

C. J. Taylor and D. J. Kriegman. Vision-based motion planning and exploration algorithms for mobile robots. In Proceedings of the Workshop on the Algorithmic Foundations of Robotics, 1994.

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