Modeling of Driver Behavior on Trajectory–Speed Decision Making in Minor Traffic Roadways With Complex Features

Abstract

The selection of trajectory and speed is a crucial factor in automobile driving behavior. However, almost all the research objects of existing Driver models are developed for driving on city streets and general highways but are not applicable to driving simulations on minor traffic roads with complex shapes, in particular. Therefore, according to the practical automobile driving processes and characteristics in the real world, this paper proposed a trajectory calculation strategy, which we named as “selecting trajectory point on a preview cross section.” According to this strategy, objective functions was established to describe the different selection patterns of “trajectory and speed” of drivers. Constraint expressions were then designed on the basis of the roadway geometry and pavement condition, passenger car performance and ride comfort. A rolling-horizon algorithm for simultaneous solving, i.e., “target trajectory–target speed,” was proposed. Eventually, validations of the proposed model were conducted using a race circuit and a complex mountain road as a simulation example, and the results indicated that the anticipated computational results could be obtained using the proposed decision-making model and algorithm. The result also showed a fairly good agreement with the trajectory and speed of a car on a practical circuit and mountain road.