Tackling the menace of fog

Tackling the menace of fog


Driving in heavy fog conditions is a major reason that leads to massive accidents. The case looks at a fog warning system that will alert drivers about poor visibility conditions and thus ensure greater safety on the road

The Abu Dhabi – Dubai Highway (E10) faced two record accidents in recent times. The first accident was in March 2008 in which more than 200 vehicles were involved in a mass collision. The second was in April 2011 and it involved 130 vehicles. These two massive accidents, and several other relatively minor ones, were due to poor visibility conditions caused by dense fog. Vehicles driving at high speed suddenly enter road sectors covered by dense fog without warning and are then implicated in mass collisions. The main objective of this system is to improve road safety in Abu Dhabi when drivers face poor visibility conditions caused by dense fog. This is achieved by sending early real-time warning signals to all drivers who are about to enter poor visibility sections of the highway, about the dangers ahead, using radio signals or cell phone voice-based short messages. Warning signals can also be displayed to the drivers using Changeable Message Signs (CMS) installed along the highway.

The proposed fog warning system consists of three components as under:

  1. Fog sensing (detection) component: This component has the main function of detecting the formation of fog and determining the geographical locations of the boundaries of foggy zones. There are several methods for determining the conditions and locations of poor visibility sectors of the highways and these include:
  1. Fog sensors: Installation of fog sensors on light poles, radar stations or cell phone towers along the highway. Data from the sensors is collected by an appropriate communications network and fed to the fog data collection and analysis component.
  2. Real-time incident detection tools/applications: Detection of slow traffic movement due to poor visibility conditions by the cell phone networks which tracks the movement of wireless devices inside vehicles driving along the highway. Such information is then passed on to the fog data collection and analysis system.
  3. Police cars on patrol: in this case, the patrol officer would approach the boundary of the fog-afflicted zone and send a signal to the data collection component while at the boundary point. This is done by simply pushing a single screen icon on a smart phone set which is programmed to send data through the 3G network. This data contains the coordinates of the boundary of the foggy zone to the fog data collection and analysis component. This technique is very efficient because it is dynamic (not restricted by the location of the sensors) and can cover any zone in Abu Dhabi area.
  1. The fog density data collection and analysis component: This component is a computational server programmed to collect real-time data from one or more of the sources of information of the fog sensing components and to conduct the appropriate analysis to determine, on the GIS of the Abu Dhabi area, with reasonable degree of accuracy the geographical boundaries of the poor visibility sectors of the highway caused by fog formation. It is important to mention that the accuracy of the virtual foggy zone depends on the number of patrol cars detecting the fog. This system saves the detected foggy zones in its database for further analysis in order to define which area of Abu Dhabi is affected frequently by fog so the decision makers can take further actions and precautions concerning this issue.
  1. Driver fog notification (signalling) component: Once the boundaries of the fog-affected zone(s) are determined, they will be sent to the server of the telecommunication operator, and this later will transmit warning signals to all drivers approaching the poor visibility sectors and well before they reach such location in order to take appropriate precautions. This is essentially the function of the driver notification (signalling) component.

In this system, there are three possible methods for driver notification:

Option 1: In this option the server sends a notification message to all the CMSs around or inside the foggy zone. The concerned CMSs will be defined by using the GIS of Abu Dhabi Area. This is the simplest and most straightforward approach. In practice, this approach has two shortcomings: the first is that the drivers may miss reading the sign at the appropriate time, and the second is the inadequate warning distance if the sign is too close or too far from the boundary of the foggy zone.

Option 2: Activating a channel on the vehicle radio that would automatically warn the drivers of the danger ahead. This will require that all vehicles are equipped with dedicated radio weather channels that are remotely activated, which is not a practical approach. Furthermore, the warning signal is neither user selective, nor location selective. That is, all drivers will receive the warning signal.

Option 3: The transmission of SMS (data or voice) to the drivers’ cell phone devices with the fog warning signal: In this case, the cell phone device will be pre-programmed to display the signal once received from the driver notification component. Only drivers approaching the boundaries of the poor visibility zone will receive such warning messages. The key issue here is that in order to receive such warning signal in time, the driver must have the cell phone switched on all the time and it must be possible to program the device so that its GIS location is tracked and it is able to identify and interpret the warning signal. Most modern cell phones have these capabilities. For subscribed users, the system can propose them the alternative way to reach their destination, if it has been already mentioned by the drivers. In the first prototype of the system, only this option will be implemented.

The proposed system is intended to overcome the major drawbacks of fog warning systems, which have been developed previously. In existing systems, fog detection is accomplished using sensors or similar devices which have specific fixed positions, while in the proposed system we benefit from the mobile and smart phone technologies in order to detect the fog dynamically in any location of the coverage area. Moreover, the detected information about fog conditions can be analysed using advanced techniques which are based on location and artificial intelligence. By using such advanced analysis techniques we define efficiently and accurately the boundaries of the fog zones in the coverage area under consideration. Concerning the notification component, the previous systems use fixed notification devices such as CMS where the locations are fixed, while in the proposed system we use a dynamic mobile and early notification technique whereby the users are warned ahead of time and before they approach the fog zone. Using advanced technologies such as GIS and mobile technologies makes our system more efficient than the existing systems. In terms of the architecture, our system can be distinguished from the previous systems because its components can be specified, designed and implemented independently. The system is thus based on a flexible, modular structure. Such a flexible structure will make it possible in the future to enhance each component without having to redesign the other components. Examples of such enhancements include incorporating new sensor technologies in future into the first component, adding more analytical capability to the second component, and adding more channels or media to the driver fog notification component to enhance reliability and improve user access. Moreover, this system can be easily configured in order to provide early warning on other weather conditions such as storms, heavy rains, etc.