摘要
针对汽车质心侧偏角难以直接获取的问题,提出了基于径向基神经网络与驾驶员-汽车闭环系统相结合的侧偏角估计方法。把汽车侧偏角看作横摆角速度和侧向加速度时间序列的映射,采用均匀设计方案对训练样本进行优选,通过神经网络建立三者之间的映射关系。同时设计了一种改进自适应卡尔曼滤波算法,将其运用到相同道路输入下汽车侧偏角的估计当中。对两种方法进行了基于实车试验的比较:神经网络方法的估计误差均值和标准差分别为0.046333°、0.057 822;°自适应卡尔曼滤波方法的估计误差均值和标准差分别为0.062 745°、0.089 241°。研究结果可以为汽车稳定性控制系统估计器的设计提供理论指导。
Aiming at the problem that vehicle side slip angles are difficult to measure directly, a radial basis function （RBF） based neural network method is proposed to estimate side slip angles combined with driver-vehicle closed-loop system. Vehicle side slip angle is considered as mapping of time series of yaw rate and lateral acceleration. A uniform design project is used to select training samples, and the relationship of the three state parameters is established through neural network. An improved adaptive Kalman filter algorithm is designed to estimate vehicle side slip angles in the Same road in- put. The two methods are compared based on full vehicle test： the average error and the standard deviation of RBF neural network method is 0. 046 333°and 0.057 822° respectively. The average error and the standard deviation of Kalman filter method is O. 062 745°and 0. 089 241° respectively. The conclusions can provide theoretic direction for design of estimator in vehicle stability control system.
出处
《南京理工大学学报》
EI
CAS
CSCD
北大核心
2009年第1期122-126,131,共6页
Journal of Nanjing University of Science and Technology
基金
高等学校博士学科点专项科研基金项目(20040287004)
关键词
汽车
侧偏角
径向基函数
神经网络
自适应卡尔曼滤波
状态估计
vehicles
side slip angles
radial basis functions
neural network
adaptive Kalman fil- ter
state estimation
