论文部分内容阅读
针对动力学建模方法对车辆质心侧偏角进行估计所面临的路面附着系数和车辆参数无法准确获取等缺点,基于统计学理论中的支持向量机对车辆质心侧偏角估计展开研究。选择方向盘转角、车辆速度、横摆角速度和侧向加速度作为支持向量机的特征向量。在Carsim仿真平台设计了20组典型车辆操纵试验作为训练样本得到预测模型,通过2组变附着系数路面上的操稳性试验对模型进行了验证。研究结果表明:支持向量机可以有效实现对不同附着路面上车辆质心侧偏角的估计,达到了较高的估计精度,即使车辆发生大侧偏现象使轮胎进入侧偏角-侧偏力曲线的非线性域,该方法仍能够实现质心侧偏角的准确估计,估计的绝对误差不超过1.42°,从而为车辆主动安全控制提供了参考。
In view of the disadvantage that the dynamic modeling method can not estimate the vehicle attachment parameters and vehicle parameters when estimating the vehicle center of mass roll angle, the vehicle center of mass side slip angle is estimated based on the support vector machine (SVM) in statistical theory. Steering wheel angle, vehicle speed, yaw rate and lateral acceleration are selected as the eigenvectors of SVM. In the Carsim simulation platform, 20 sets of typical vehicle maneuvering tests were designed as training samples to get the predictive model. The model was verified by two sets of stability tests on the pavement with variable coefficients. The results show that the support vector machine can effectively estimate the vehicle center of mass offsets on different attached roads, and achieve a high estimation accuracy. Even if the vehicle has a large side-slip phenomenon to make the tire enter the side-slip angle-lateral deflection curve Nonlinear domain, this method can still achieve accurate estimation of the center of mass side slip angle, the absolute error of estimation does not exceed 1.42 °, which provides a reference for the active safety control of vehicles.