The Driver Model Under Complex Driving Conditions

Transportation Safety has been a social problem with the popularization of automobile and the development of automobile technology. The development of intelligent vehicles traffic and the automobile active safety have been paid more and more attention. Throughout the research, as the leading factors of the automobile active safety, the vehicle handling stability and the driver's behavior property play an important role in it. At the same time, these two ones are interactively connected within the driver-vehicle close loop system. It has been proved that considering the relationship among the driver, vehicle and the road as a whole and to investigating the vehicle handling stability from the systematic aspect is on the right tack, both of which had been made possible by the development of the driver model, which has importance significance on the study of the closed-loop evaluation of vehicle handling and stability and intelligent traffic especially intelligent vehicles.The objective of this thesis is to establish effective, accuracy and simple driver model under complicated trajectories, and to embed the driver model in dynamic software CarSim which is widely used in automotive industry, and to simulate the driver-vehicle closed-loop system under typical trajectories. By doing this, the time for application of Driver model in vehicle dynamics simulation is ripe. Firstly, according to the actual application of the driver model, basing on the Preview Optimal Artificial Neural Network (POSANN) Driver Model and the Error Elimination Algorithm (EEA) in driver modeling, this paper embeds the UniTire and the driver model in complicated vehicle, and consummate directional control driver model. And the driver model is applied in complicated vehicle with variational speeds especially low speed.Secondly, based on OPA method, this paper develops Direction control and Speed Control Combined Driver Model for OPA method and POANN method. The preview point searching algorithm is proposed to make the driver model be able to follow arbitrary designated path and speed defined in ground coordinate. Driver/vehicle closed-loop simulations indicate that the proposed driver model is suitable for arbitrary path and speed following at varying vehicle speed.Lastly, under emergency condition (which occur before traffic accident), environment of the road and property of vehicle changed, the dynamics property of vehicle changed suddenly. At the same time the transfer function of vehicle has difference with desired property of this vehicle. The driver didn't adapt this change immediately, the driver was difficult to control this vehicle, the traffic accident will happen. The change of vehicle transfer function has important significance on the safety of vehicle under the emergency condition. Considered the behavior of actual driver driving vehicle, we improved the driver model under the emergency condition. Based this idea, we predigest the parameters of the driver model and make the driver model has adaptability on the property of vehicle and environment of the road under emergency condition. Discussion in the paper is the application of UniDriver Model, which based on modifying of lateral acceleration, in the complicated vehicle model of CARSIM and the model is proved to be able to adequately describe the driver's performance characteristic by simulation under various conditions.In a conclusion, driver modeling is the essence of the ongoing development of intelligent vehicles as well. Aiming at replacing the expert human driver with the simulated driver model completely or partially, intelligent vehicles are designed to run with efficiency and safety, which call for the in-depth understanding of the human driver behavior and the reliable constitution of the driver model. This proves that the driver model is quite reliable and efficient to simulate the human driver steering behavior.