线控转向系统的神经网络模型与模糊控制

采用魔术公式表达的轮胎模型,建立线控转向系统非线性三自由度整车动力学模型,该模型在整个轮胎侧偏角范围内有效,反映轮胎侧偏角超过5°后轮胎侧偏力与轮胎侧偏角的非线性特性,并采用BP神经网络训练模型。转向传动比采用模糊控制。结果表明：采用BP神经网络训练模型较好的映射了输入与输出的非线性关系,采用转向传动比模糊控制算法得到转向传动比随车速的变化规律,可较好的实现低速转向灵敏和高速转向稳定的控制目标。

On Neural Network Model and Fuzzy Control of Steering-by-Wire System

In this article,a nonlinear overall vehicle dynamics model with the three-degree freedom is established by using the tire model based on Magic Formula,which is effective in the whole tire slip angles and reflects the nonlinear characteristics between tire lateral forces and tire slip angles while the tire slip angles are larger than five degrees. Then,the model is trained by the BP neural network. Finally,a fuzzy control is adopted for the steering ratio. The results show that the BP neural network training of vehicle model reflects the nonlinear relations between the inputs and the outputs well. The steering ratio changing with velocities by using the fuzzy control algorithm can realize the control goals of both steering agile under low velocities and steering stability under high velocities.