Identification of Traffic States From Onboard Vehicle Sensors

This paper describes an algorithm that identifies the state of traffic ahead of a moving vehicle using onboard sensors. This algorithm approximates the level of service as defined in the Highway Capacity Manual, which portrays a range of traffic conditions on a particular type of roadway facility. The traffic state forms an independent variable in an evaluation plan to assess the benefits and capability of an automotive rear-end crash avoidance system in a field operational test. The algorithm utilizes inputs from vehicle sensors, onboard radar, global positioning system, and digital map to classify the traffic ahead into light, medium, and heavy states. Basically, the algorithm segregates the roadway into four different categories based on the road type (freeway or non-freeway), posted speed limit, and traffic flow conditions. In addition, the algorithm computes two key parameters: (1) number of vehicles in the radar field of view that are moving in the same direction as the test vehicle, and (2) speed ratio between the vehicle travel speed and the posted speed limit. A logical process that ties these two parameters to each of the four roadway categories then determines the traffic state. The conditions of the logical process were optimized by minimizing the error between the calculated and observed traffic states using samples of video and numerical data collected from an instrumented vehicle on public roads.