Abstract: To address the oscillation issue in electrical power system (EPS) caused by external interference during intelligent drive mode, this paper conducts detailed analysis and identifies insufficient robustness of the proportional-integral-derivative (PID) controller responsible for steering angle closed-loop control in EPS as the root cause of oscillation phenomenon. Through force analysis for mechanical steering system, differential equations of motions are established and transformed into the form of state equations. By defining performance indicator, an linear quadratic regulator (LQR) optimal controller is constructed. The weighting matric coefficient of LQR optimal controller is determined via simulation analysis. The vehicle tests are conducted as well and the results show that, compared to the PID controller, the LQR optimal controller demonstrates superior robustness and tracking performance, and it not only completely resolves the oscillation issue in EPS but also achieves a certain degree of improvement in the key indicators of tracking performance. This paper also provides a practical approach for steering angle control in steering-by-wire system.

Key words: intelligent drive; PID; LQR; oscillation phenomenon

摘要： 为了解决智能驾驶模式下电动助力转向系统（EPS）受到外界干扰而产生的抖动问题， 文章经过详细分析，发现 EPS 中负责转角闭环控制的比例-积分-微分（PID）控制器的鲁棒性 不足是产生抖动现象的根本原因。通过对机械转向系统的受力分析，建立其运动微分方程， 并转换为状态方程的形式，给定性能指标，建立线性二次型调节器（LQR）最优控制器。通 过仿真分析确定了 LQR 最优控制器的加权矩阵系数，并进行了实车验证。结果表明，相对于 PID 控制器，LQR 最优控制器表现出更好的鲁棒性和跟随性，不但完全解决了 EPS 的抖动问 题，而且跟随性能的主要指标也有一定程度的提升。文章也为线控转向的角度控制提供了一 种切实可行的方式。

关键词: 智能驾驶；PID；LQR；抖动现象