A new car-following model considering driver’s sensory memory

• A new memory car-following model is proposed.
• The memory effects on the stability of traffic flow have been investigated.
• The new car-following model compensates for the disadvantage of the sensory buffer time neglected in existing models.
• The new memory car-following model can convey following car’s driver sensing behavior better and avoid the negative velocity in existing models.

This paper presents one kind of new car-following model (mean memory model, simplified as MMM) by introducing driver sensory memory (sensory buffer) term into the original optimal velocity (OV) function by Bando et al. (1995, 1998). The main improvement is that MMM can avoid the disadvantage of the sensory buffer time neglected in existing models. The stability condition of the proposed model is obtained by using linear stability theory. Results show that the stability region decreases when the driver’s sensory buffer time increases. Furthermore, the model is investigated in detail by numerical methods. The following conclusions are derived. (a) Numerical results of starting process for the car motion under a traffic signal accord with empirical traffic values; (b) the numerical simulations in the form of the space–time evolution of headway and velocity are also in good agreement with the theoretical analysis; (c) the size of hysteresis loops will be reduced with the sensing buffer time decreasing. Both analytical and simulation results show that the following car driver’s sensory buffer time plays an important role on the stability of traffic flow.