Jam Tail Warnings Based on Vehicle Probe Data

Abstract

Abstract Recent research shows every so often upcoming cases of dangerous accidents happening at jam tails so that the introduction of a proper warning system is helpful to alleviate the problem. Since traffic is a highly dynamic and complex system of spatiotemporal patterns one has to have an outright understanding of the underlying phases of the real traffic to make a qualified assertion when a dangerous jam tail in a traffic pattern occurs. Kerner showed with the three-phase traffic theory an alternate approach of traffic modeling to the classical theories based on the fundamental diagram. With the help of splitting traffic into three distinct phases: free flow, synchronized flow and wide moving jams, potentially dangerous situations can be detected and identified. A method of identifying critical jam tails based on microscopic vehicle probe data and the three-phase traffic theory is proposed. With the sheer amount of vehicle probe data available today a method to identify jam fronts and detect critical jam tails has been developed and evaluated with vehicle probe data. The stable propagation of dangerous jam tails and their detection based on vehicle probe data is presented for a freeway example in the UK. It is further discussion if and how automated and autonomous vehicles can use this information to adapt the behavior to a more safe and comfortable assisted braking and increased safety at jam tails.