Cooperative Situation Awareness in Transportation

Intelligent Transportation Systems (ITS) became a fast moving field of research in the last decades, in particular in the context of continuously growing mobility and a high employment of resources starting from energy and material consumption to travel time and finally the human life. As it has already been experienced in other application areas, the introduction of communications technology is able to bring a revolutionary change in structures and behaviors long-believed to be carved in stone. The main idea behind this thesis is the usage of information not as a mere placeholder, e.g. a field in a static message, but actively utilizing its content and dependencies. This requires an estimation of the actual worth of a single piece of information for the entity itself and the entities which are in communication range. This worth has to be the essential driver for the cooperative situation estimation. The active utilization of information and its cooperative dissemination provides the entities the opportunity to become situation aware and detect hazardous or inefficient situations early in advance. Since information always has a degree of uncertainty which is inherent to information in the real-world problem domain, as we are confronted with in ITS, probabilistic methods will be applied to model situation-relevant information. Conditional probability distributions in state transition models make for the evolvement of the situational information with the progress of time and handle causal dependencies between situational information. Together with a utility-based decision-making process dynamic probabilistic causal decision networks provide the functionality to select optimal actions given sequences of inaccurate and incomplete evidences. This thesis provides concepts and strategies that push forward the exploitation of information in a cooperative way within a probabilistic framework that allows to make various kinds of decisions with maximum utility. For the evaluation of the proposed concepts, the exemplary application Cooperative Adaptive Cruise Control (CACC) has been implemented on the basis of a particle filter which is used for the situation estimation. Initial simulations provided promising results and hence constitute a solid basis for future work in the field of Cooperative Situation Awareness in Transportation.