基于MPC的驾驶员转向控制行为建模与仿真

Modeling and Simulation of Driver Steering Behavior Based on Model Predictive Control

为了描述驾驶员的转向行为,研究了驾驶员转向过程中的预瞄及滚动优化行为,由于难以用简单的数学表达式反映汽车转向过程中驾驶员的行为,提出选用搜索算法及模型预测控制方法(Model Predictive Control,MPC)进行具体建模。最后,结合车辆动力学软件veDYNA对所建立的驾驶员转向行为模型进行验证。仿真结果表明,驾驶员转向控制行为模型对人类驾驶员的仿真相似程度较高,符合人类驾驶员的驾驶特点,说明改进模型预测控制理论为研究驾驶员模型提供了有效途径。

To imitate the behavior of a driver steering, the abilities of preview, cognition, prediction and optimi- zation were researched. Because the driver＇ s behavior is complex in the process of vehicle steering, it is difficult to be expressed with a simple mathematical. Sdecting the search algorithm and Modal Predictive Controi, the driver steering behavior was modeled. At last, combined with vehicle dynamics software veDYNA, the simulation was con- ducted with the real vehicle data. Compared with the results of the human driver, the driver steering behavior model has a high similarity, and can meet the human driver＇s driving characteristics. Therefore, the model predictive control theory can provide an effective way for driver model study.