| Summary: | 碩士 === 國立臺南大學 === 數位學習科技學系碩士班 === 97 === Traffic information plays an important role in vehicular ad hoc networks (VANETs). It provides the information among vehicles and surrounding circumstances on the roads. Common traffic information includes mean speed of road segments, mean density of road segments, traffic flow, and traveling time estimation. If above traffic information could be used properly, some traffic problems which are like traffic congestion and traffic accident may have released. However, through the traffic flow theory, only the speed or density, even traffic flow, can not accurately judge traffic conditions on the roads. Therefore, the method Acquiring Confidential Traffic Information Of Need (ACTION) is proposed to model real time traffic information. By ACTION, the relationship between speed and density through clustering and modeling is transferred as a novel parameter of traffic, height, to express real-time traffic conditions based on traffic flow theory. In addition, the collections of traffic information are still estimated by the fixed equipment, and are subject to weather and other natural factors. In order to take into account the timeliness of data and the accuracy, data collection of this thesis is through vehicular ad hoc networks. Finally, the traffic information is applied into navigation of vehicles as an example. In the method of navigation, road network topology is seen as an artificial potential field, each intersection get a virtual force computed by the gradient operator in all directions. When vehicles send packets to roadside units (RSUs), the system will guide vehicles after setting the destination. The simulation results show that traffic conditions are expressed well by the modeled parameter proposed in this paper, and using the methods of navigation by virtual forces can be effective ways to lead vehicles avoid traffic congestion.
|
|---|
