A fuzzy logic approach for car-following modelling

In this communication, the classical car-following theory, which is based on intuitiveness and behavioural representation, is revisited from the point of view of cybernetics theory. Then, a new approach, based on fuzzy modelling, is proposed to estimate the main parameters of the basic car following model. This approach, contrarily to earlier fuzzy logic approaches, maintains the structure of the classical car-following model, limits its complexity and provides means to take into account the main parameters influencing the behaviour of the driver and allows a formal verification of local and global stability of the simulated traffic flow. So, after introducing a new approach based on cybernetics concepts for the modelling of car-following, which provides a theoretical basis for the basic model, the main concepts of fuzzy logic theory and of fuzzy modelling are briefly presented. Then a fuzzy estimation approach is proposed for the main parameters of the basic car-following model. Finally, simulation results making use of the proposed modelling approach are presented.