A predictive accident-duration based decision-making module for rerouting in environments with V2V communication
Providing drivers with relevant information about the environment surrounding their vehicle can assist them in making driving safer and easier. This information can be useful to the driver, but there is also the possibility of overwhelming the driver. Some mechanisms already exist to help the driver act in these situations. However, vehicles can also send messages in emergency situations, such as when a crash occurs, to warn other vehicles. In addition to increasing safety, this information can be helpful in decreasing traffic congestion near a crash location. This paper describes an on-board vehicular ad-hoc network (VANET) based decision-making module that receives crash information from other vehicles, informs the driver about it, and suggests an alternative route to avoid crash-related congestion.
The on-board VANET decision-making module was implemented and tested with vehicles in network simulation (Veins), which uses OMNet++ (a wireless network simulation tool), linked to SUMO (road network simulation software). Testing was conducted on an approximately 4 km by 4 km road network of Erlangen, Germany and Highway 401 in Ontario, Canada. Different simulations also considered different rates of module adoption, ranging from 10 to 100 percent.
The decision-making system uses a crash model to predict the duration of a crash based on factors including location, number of lanes of the facility, number of lanes blocked, and the type of vehicles involved. Given this information, the module proposes an alternative route for vehicles approaching the crash scene to avoid crash-related congestion.
The module uses an event-based decision-making approach for vehicle rerouting and triggers only when crash messages are received to avoid redundant messaging. The system also works between any equipped vehicles and does not require a central management control mechanism.
In urban environments, both travel and waiting times are reduced with increasing adoption rates. On highways, the system could yield less waiting time in lieu of a slightly increased travel time. The following tables summarize the approximate differences in waiting times at different levels of adoption. During heavy traffic periods, waiting time was reduced by approximately 80 percent in urban areas and 42 percent on freeways with complete adoption of the module.
Approximate Waiting Time in an Urban Environment (mins)
Approximate Waiting Time in a Freeway Environment (mins)