No abstract

Recent years have witnessed a rapidly growing interest in using teams of mobile robots for autonomously covering environments. In this paper a novel approach for multi-robot coverage is described which is based on the principle of pheromone-based communication. According to this approach, called StiCo (for "Stigmergic Coverage"), the robots communicate indirectly via depositing/detecting markers in the environment to be covered. Although the movement policies of each robot are very simple, complex and efficient coverage behavior is achieved at the team level. StiCo shows several desirable features such as robustness, scalability and functional extensibility. Two extensions are described, including A-StiCo for dealing with dynamic environments and ID-StiCo for handling intruder detection. These features make StiCo an interesting alternative to graph-based multi-robot coverage approaches which currently are dominant in the field. Moreover, because of these features StiCo has a broad application potential. Simulation results are shown which clearly demonstrate the strong coverage abilities of StiCo in different environmental settings.

n this chapter, we study the application of existing entity resolution (ER) techniques on a real-world multi-source genealogical dataset. Our goal is to identify all persons involved in various notary acts and link them to their birth, marriage, and death certificates. We analyze the influence of additional ER features, such as name popularity, geographical distance, and co-reference information on the overall ER performance. We study two prediction models: regression trees and logistic regression. In order to evaluate the performance of the applied algorithms and to obtain a training set for learning the models we developed an interactive interface for getting feedback from human experts. We perform an empirical evaluation on the manually annotated dataset in terms of precision, recall, and F-score. We show that using name popularity, geographical distance together with co-reference information helps to significantly improve ER results

Visual feature detection with limited resources of simple robots is an essential requirement for swarm robotic systems. Robots need to localize their position, to determine their orientation, and need to be able to acquire extra information from their surrounding environment using their sensors, while their computational and storage capabilities might be very limited. This paper evaluates the performance of an experimental framework, in which environmental elements such as landmarks and QR-codes are considered as key visual features. The performance is evaluated for environmental light disturbances and distance variations and feature detection speed is thoroughly examined. The applicability of the approach is shown in a real robot scenario by using epuck robots. Finally, the results of applying the approach to a completely different setting, i.e., simulation of pheromones using glowing trail detection, are presented. These results indicate the broad applicability range of the developed feature detection techniques.