Analysis of risk levels for traffic on a multi-lane highway

Abstract

We present an analysis of risk levels on multi-lane roads. The aim is to use the crash metrics to understand which direction of the flow mainly influences the safety in traffic flow. In fact, on multi-lane highways interactions among vehicles occur also with lane changing and we show that they strongly affect the level of potential conflicts. In particular, in this study we consider the Time-To-Collision as risk metric and we use the experimental data collected on the A3 German highway.

A two-dimensional data-driven model for traffic flow on highways

“…See Figure 2. In [36] we also show, using the same German data-set described above, that the level of potential conflicts is strongly affected by the lane changing. To this end we computed the average risk on the two directions of the flow by means of the Time-To-Collision metric and we observe that the risk due to lane changing is higher.…”

Crowd Dynamics: Modeling and Control of Multiagent Systems

“…This model assumption has been motivated by a study showing that only 15% of vehicles change lanes while traveling along a road section and that the chance of two vehicles performing a lane change at exactly the same time is negligible (112).…”

A Holistic Approach to the Energy-Efficient Smoothing of Traffic via Autonomous Vehicles

“…This constraint is motivated by two reasons: the lane change is not instantaneous (so there should be a fixed minimal duration between two lane-changes), a driver that has just changed lanes is less likely to change lanes again right away. From [24], only the 15% of the vehicles cross a lane while traveling the road section. In the simulation presented in [19], we choose the cool-down time T 1 = 5s over the time interval [0, 1000s].…”