引入置信度方法的变参数路面附着系数非线性观测器

Road adhesion state observer with adaptive parameter based on confidence coefficient method

为了提高观测器估计精度并针对大多数采用固定观测器增益的现状,根据Lyapunov稳定性条件采用变参数方法,实现观测器增益的自适应调整。目前大多数观测器估计值为利用附着系数并不是峰值附着系数,因此引入置信度的方法,即通过带补偿度的神经网络模糊算法计算置信度从而估算峰值附着系数。采用Simulink与Carsim动力学仿真软件进行联合仿真验证,结果表明设计的非线性观测器是有效的,估计精度满足工程要求。

Firstly in order to improve road adhesion sate estimation accuracy,the fixed observer gain that most reference articles apply was modified into adaptive gain based on Lyapunov stability limitation.Secondly the estimated value of most observers was the used but not the peak,the combination of confidence coefficient and the compensatory neuronfuzzy algorithm(CNFA) was applied to estimate the peak value.Finally co-simulation of Simulink and Carsim was implemented,simulation results validate the effectiveness of the integration of confidence coefficient and CNFA,and nonlinear observer estimation accuracy meets engineering specification.