Agent Based Sensor and Data Fusion in Forest Fire Observer

Abstract

Although natural hazards have been present on earth forever, we still have not developed effective way of dealing with them, and because of that the early detection of natural hazards is still quite important task. Traditional methods based on human observers are mostly used, but they have shown ineffective because of a human observer’s subjectivity. Modern technologies, especially sensors technology provide tools for new techniques of natural hazard early detection. The sensor network technology is one of the most important technologies currently being investigated by scientist all around the world. The sensors, creating a sensor network, today could be deployed in nature on places where it was not possible to put sensors before, particularly because of today’s development of wireless communication and miniature autonomous power supplies. Collecting data from sensors creating a sensor network is one part of the research in this field, and another one is sensor network data utilization. The goal of sensor data interpretation, particularly using data fusion techniques, is to understand the sensor network surrounding and to create the image of the world around the sensors. In this paper this image of the world around the sensors will be call the scenario of the phenomenon. Combining sensors for collecting data and advanced algorithms for data processing and interpretation, more advanced object called the observer could be designed. Combining several observers an observer network could be established, capable for better understanding what is going on in sensors surroundings. In such a way an advanced early warning system for initial phase natural hazards detection could be designed. When done by humans, the process of recognition the phenomenon of interest based on fusion of sensory inputs is called perception. Our work was primarily inspired by formal theory of perception introduced in 1989 by Benett, Hoffman and Prakash (Bennett at al., 1989). This theory is used as a framework for formal description of data fusion processes in a observer network. The observer network is a system consisting of its physical part, various sensors deployed directly in the environment integrated with appropriate communication components, and software part, intelligent mechanisms for sensor network data fusion and interpretation. Data fusion is used as a formal framework for combining data from different physical sensors of the same type, but also from different type of sensors. Data fusion main task is to obtain information of greater quality. The exact definition what ‘greater quality’ means is primarily application dependent (Wald, 1999). O pe n A cc es s D at ab as e w w w .in te ch w eb .o rg