Nowadays, among the microscopic traffic flow modeling approaches, the car-following models are increasingly used by transportation experts to utilize appropriate intelligent transportation systems. Unlike previous works, where the reaction delay is considered to be fixed, in this paper, a modified neural network approach is proposed to simulate and predict the carfollowing behavior based on the instantaneous reaction delay of the driver-vehicle unit as the human effects. This reaction delay is calculated based on a proposed idea, and the model is developed based on this feature as an input. In this modeling, the inputs and outputs are chosen with respect to the reaction delay to train the neural network model. Using the field data, the performance of the model is calculated and compared with the responses of some existing neural network car-following models. Considering the difference between the responses of the actual plant and the predicted model as the error, comparison shows that the error in the proposed model is significantly smaller than that that in the other models.

No abstract

Nowadays, vehicles are the most important means of transportation in our daily lifes. During the last few decades, many studies have been carried out in the field of intelligent vehicles and significant results on the behavior of car-following and lane-change maneuvers have been achieved. However, the effects of lane-change on the car-following models have been relatively neglected. This effect is a temporary state in car-following behavior during which the follower vehicle considerably deviates from conventional car-following models for a limited time. This paper aims to investigate the behavior of the immediate follower during the lane-change of its leader vehicle. Based on a closer inspection of the microstructure behavior of real drivers, this temporary state is divided into two stages of anticipation and evaluation. Afterwards, a novel and adaptive neuro-fuzzy model that considers human driving factors is proposed to simulate the behavior of real drivers. Comparison between model results and real traffic data reveals that the proposed model can describe anticipation and evaluation behavior with smaller errors. The anticipation and evaluation model can modify current car-following models so as to accurately simulate the behavior of an immediate follower which leads to an enhancement of carfollowing applications such as driving assistance and collision avoidance systems.

No abstract

Car-following and lane-changing manoeuvres are the most common driving behaviour on urban roads and highways. Although these two manoeuvres have been studied extensively, the effect of a lane change on a car-following manoeuvre remains elusive. Analysing these effects leads to integration of the car-following and the lane-changing manoeuvre which has been relatively neglected. A lane-changing manoeuvre causes the immediately following driver to deviate from common car-following models to accommodate the lane changer ahead; this is called anticipation and relaxation behaviour. These behaviours are transient states which occur between two car-following behaviours owing to the lane-changing manoeuvre. In this paper, a novel adaptive neurofuzzy model is proposed for simulating the behaviour of the follower vehicle during anticipation and relaxation behaviour. Comparison between the simulation results and the field data shows that errors in the proposed model are significantly smaller and the model can describe anticipation and relaxation behaviour properly. The anticipation and relaxation model can improve current car-following model applications to enhance the safety of vehicles such as driving assistant and collision avoidance systems.