Fuzzy Traffic Density Homogenizer for Automated Highway Systems

A major factor that contributes to congestion on today’s freeways is the severe inhomogeneities of the traffic streams. If the traffic density can be controlled, congestion can be eliminated and delays can be reduced. This goal can be achieved by properly controlling the mean speed of traffic on each section of the freeway lane as well as the ramp entering rates. In this paper a fuzzy traffic density homogenizer is proposed to alleviate or avoid congestion by smoothing the traffic density distribution profile over the freeway lanes. The proposed fuzzy traffic density homogenizer consists of two parts: the fuzzy mean speed controller and the fuzzy on-ramp controller. The fuzzy mean speed controller is designed to determine the desired mean speed to be followed by vehicles in each section in order to achieve smooth traffic flow. The fuzzy on-ramp controller is designed to determine the maximum allowable on-ramp traffic volume in each section so that the on-ramp traffic flow does not lead to an amplification of congestion. Simulation results are used to demonstrate that the proposed controllers can eliminate traffic flow instabilities leading to smooth traffic flows.