The invention discloses a vehicle state estimation method and system combining a neural network and unscented Kalman filtering, and the method comprises the steps: building the neural network through the data obtained from a plurality of sensors of an automobile, and precisely estimating the three-way acceleration and three-way angular velocity. The state can define the movement of the chassis of the automobile so as to realize the automobile state required for coordinating the subsystems. The 3D attitude, speed and positioning of the automobile can be better estimated by combining unscented Kalman filtering of the neural network. According to the invention, accurate and comprehensive vehicle dynamic information is provided. Strong coupling caused by signal exchange among all parts of the vehicle is prevented, and therefore the accuracy of a single state estimation result is prevented from being affected. The vehicle state is accurately estimated by adopting the neural network for state prediction, and compared with a past neural network state estimator, the scheme is easier to train and more accurate. Based on the output of the neural network, the UKF can better estimate the 3D attitude, speed and positioning of the automobile. When the GPS signal is unavailable, the method can provide high-precision navigation information.

    本发明公开了一种结合神经网络与无迹卡尔曼滤波的车辆状态估计方法及系统,利用汽车多传感器中获得的数据,建立神经网络精确估计三向加速度与三向角速度。该状态可以定义汽车底盘的运动以实现协调各子系统所需要的汽车状态。结合神经网络的无迹卡尔曼滤波可以更好的估计汽车的3D姿态、速度、定位。本发明提供准确、全面的车辆动态信息。防止车辆各部件之间的信号交换导致的强耦合,从而影响单个状态估计结果的准确性。采取分状态预测的神经网络准确估计汽车状态,相较于过去的神经网络状态估计器,该方案更易训练,更准确。基于神经网络的输出,UKF能更好的估计出汽车3D姿态、速度、定位。当GPS信号不可用时,本方法能提供高精度的导航信息。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Vehicle state estimation method and system combining neural network and unscented Kalman filter


    Weitere Titelangaben:

    一种结合神经网络与无迹卡尔曼滤波的车辆状态估计方法及系统


    Beteiligte:
    CAI YINGFENG (Autor:in) / TAN CHENKAI (Autor:in) / CHEN LONG (Autor:in) / LIU WEIGUO (Autor:in) / ZHOU JUNJIE (Autor:in) / WANG HAI (Autor:in) / SUN XIAOQIANG (Autor:in)

    Erscheinungsdatum :

    2023-01-03


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion



    Unscented Kalman filter for vehicle state estimation

    Antonov,S. / Fehn,A. / Kugi,A. et al. | Kraftfahrwesen | 2011


    Unscented Kalman filter for vehicle state estimation

    Antonov, S. | Online Contents | 2011


    Unscented Kalman filter for vehicle state estimation

    Antonov, S. / Fehn, A. / Kugi, A. | Taylor & Francis Verlag | 2011


    Vehicle State Information Estimation with the Unscented Kalman Filter

    Ren, Hongbin / Chen, Sizhong / Liu, Gang et al. | Tema Archiv | 2014


    Unscented Kalman Filter

    Zarchan, Paul / Musoff, Howard | AIAA | 2015