Research On Driver Steering Control Model Based On Visual Behavior When Driving In The Crooked Roads

As the core control unit of decision-making in the closed-loop system of driver-vehicle-road,the driver steering control modeling is of great significance for the vehicle control stability and the decision control of intelligent driving system.Drivers often obtain information to complete steering control through multiple channels,and visual information is the most important information source of driver's multi-channel information.Therefore,it is necessary to build the driver steering control model based on the visual behavior.Vehicle dynamics modeling and verification,road modeling and visual behavior modeling have been completed successively.Finally,the driver steering control model based on visual behavior has been established by using fuzzy control theory.Simulation experiments have verified that the established model has good stability and path following ability.The main research content is as follows:Firstly,considering that the driver steering control model is based on the closed-loop system of driver-vehicle-road,the vehicle,road and driver were modeled respectively.Based on the control stability,a two-degree-of-freedom vehicle dynamics model was established to simplify the vehicle system,and its agreement with the CarSim dynamics model was verified.The piecewise function and the seventh degree polynomial were used to fit the road trajectory respectively,and the double-lane change and double S curve road model were established.Based on the driver's visual behavior on the curve,a two-point preview model was established based on far point and near point.Secondly,the matching mechanism between driver's visual behavior and steering behavior was studied to set up the decision-making model based on fuzzy controller.The driver is a complex nonlinear dynamic system,so it is difficult to establish an accurate mathematical model.Therefore,fuzzy control theory was used to model the process from visual information acquisition to steering execution,and a fuzzy controller was designed with near-point lateral deviation and preliminary steering wheel angle as input and optimal steering wheel angle as output.Finally,in order to verify the effectiveness of the driver steering control model based on fuzzy control theory and integrating the driver's visual behavior characteristics,the co-simulation analysis was conducted based on Matlab/Simulink and CarSim.The vehicle speed was 60 km/h for medium speed and 100 km/h for high speed respectively.By comparing the simulation results of the established model with the single-point preview model and the multi-point preview model,it was found that,the driver steering control model based on the driver's visual behavior and fuzzy control theory has improved the track tracking ability,as well as the steering stability and comfort.