Autonomous place detection has long been a major hurdle to topological map-building techniques. Theoretical work on topological mapping has assumed that places can be reliably detected by a robot, resulting in deterministic actions. Whether or not deterministic place detection is always achievable is controversial; however, even topological mapping algorithms that assume non-determinism benefit from highly reliable place detection. Unfortunately, topological map-building implementations often have handcoded place detection algorithms that are brittle and domain dependent.

IEEE802.11b wireless Ethernet isbu19803 the standard for indoor wireless communication. This paper proposes the use of measured signal strength of Ethernet packets as a sensor for a localization system. We demonstrate that off-the-shelf hardware can accuratelyb used for location sensing and real-time trackingb applying a Bayesian localization framework.

Abstract not found

Abstract. We present an incremental algorithm for constructing and reconstructing Generalized Voronoi Diagrams (GVDs) on grids. Our algorithm, Dynamic Brushfire,

Recent advances in many multi-discipline technologies have allowed small, low-cost fixed wing unmanned air vehicles (UAV) or more complicated unmanned ground vehicles (UGV) to be a feasible solution in many scientific, civil and military applications. Cameras can be mounted on-board of the unmanned vehicles for the purpose of scientific data gathering, surveillance for law enforcement and homeland security, as well as to provide visual information to detect and avoid imminent collisions for autonomous navigation. However, most current computer vision algorithms are highly complex computationally and usually constitute the bottleneck of the guidance and control loop. In this paper, we present a novel computer vision algorithm for collision detection and time-to-impact calculation based on feature density distribution (FDD) analysis. It does not require accurate feature extraction, tracking, or estimation of focus of expansion (FOE). Under a few reasonable assumptions, by calculating the expansion rate of the FDD in space, time-to-impact can be accurately estimated. A sequence of monocular images is studied, and different features are used simultaneously in FDD analysis to show that our algorithm can achieve a fairly good accuracy in collision detection. In this paper we also discuss reactive path planning and trajectory generation techniques that can be accomplished without violating the velocity and heading rate constraints of the UAV.

Abstract — Tasks such as street mapping and security surveillance seek a route that traverses a given space to perform a function. These task functions may involve mapping the space for accurate modeling, sensing the space for unusual activity, or processing the space for object detection. With a prior map, an optimal path can be computed using a graph to represent the environment and generating the solution using known graph algorithms. However, the prior map may be inaccurate due to factors such as occlusion, outdatedness, dynamic objects, and resolution limitations. In this work, we address the NP-hard problem of optimal environmental coverage with incomplete prior map information. To utilize related algorithms in graph theory, we represent the environment as a graph. Using this representation, we present a graph coverage approach for optimal plan generation based on the Undirected Chinese Postman and Rural Postman problems. This approach produces a tractable solution through the use of low complexity algorithms in a branch-and-bound framework. Additionally, as the robot receives sensor updates during traversal of the environment, we update the graph to reflect those changes. The updated graph can be highly disconnected so computing an optimal solution can be NP-hard. To combat this, we introduce a heuristic for coverage path generation that helps maximize the connectivity of the updated graph. We evaluate our approach on a set of comparison tests in simulation. I.

Abstract — This paper introduces a sensor-based planning algorithm that uses less sensing information than any others within the family of bug algorithms. The robot is unable to access precise information regarding position coordinates, angular coordinates, time, or odometry, but is nevertheless able to navigate itself to a goal among unknown piecewiseanalytic obstacles in the plane. The only sensor providing real values is an intensity sensor, which measures the signal strength emanating from the goal. The signal intensity function may or may not be symmetric; the main requirement is that the level sets are concentric topological circles. Convergence analysis and distance bounds are established for the presented approach. I.

We consider the problem of discovering a smooth unknown surface S bounding an object O in R 3. The discovery process consists of moving a point probing device in the free space around O so that it repeatedly comes in contact with S. We propose a probing strategy for generating a sequence of surface samples on S from which a triangulated surface can be generated that approximates S within any desired accuracy. We bound the number of probes and the number of elementary moves of the probing device. Our solution is an extension of previous work on Delaunay refinement techniques for surface meshing. The approximating surface we generate enjoys the many nice properties of the meshes obtained by those techniques, e.g. exact topological type, normal approximation, etc.

A key prerequisite for higher-level fusion is the use of context to disambiguate and interpret sensed data and guide data collection. For ground operations terrain information supplies an important context. The layout of terrain is a determining factor in arraying of forces, both friendly and enemy, and the structuring of Courses of Action (COAs). For example, key terrain, such as a bridge over an un-fordable river, or terrain that allows observation of the opposing forces line of advance, is likely to give a big military advantage to the force that holds it. Combining information about terrain features with hypotheses about enemy assets can lead to inferences about possible avenues of approach, areas that provide cover and concealment, areas that are vulnerable to enemy observation, or choke points. Key terrain identifies areas where intelligence collection effort should be focused. In addition, if force movements are observed, terrain features give additional information with respect to the intent of enemy forces that have been observed on the move, thus confirming or disconfirming hypotheses about enemy intent. Currently, intelligence officers manually combine terrain-based information, information about the tactical significance of certain terrain features as well as information regarding enemy assets and doctrine to form hypotheses about the disposition of enemy forces and enemy intent. In this paper, we present a set of algorithms and implemented tools for automating terrain based information fusion and inference. The products of automated terrain analysis are currently being validated using analysis of the same terrain produced by human intelligence officers.

Summary. Self-localization is an important task for humans and autonomous robots as it is the basis for orientation and navigation in a spatial environment and for performing mapping tasks. In robotics, self-localization on the basis of monomodal perceptual information has been investigated intensively. The present chapter looks at self-localization in a more general setting where the reference information may be provided by different types of sensors or by descriptions of locations under a variety of conditions. We introduce some of these conditions and discuss general approaches to identifying locations in perceived environments. Taking into account cognitive considerations, we propose an approach to identify locations on a high, abstract level of representation. The approach combines qualitative and quantitative information to recognize locations described as configurations of shape features. We evaluate this approach in comparison to other approaches in a selflocalization task and a generalized localization task based on a schematic map. 1