论文部分内容阅读
为了解决一般状态增强型卡尔曼滤波和两级卡尔曼滤波在微机械惯性传感器不确定噪声的影响下难以获得良好定位性能的问题,提出一种新型的基于交互式多模型的两级卡尔曼滤波方法来适应微机械惯性传感器的不确定噪声.根据不同的噪声特性,建立3个偏差滤波器来覆盖大范围的噪声水平.交互式多模型算法根据3个偏差滤波器可以准确估计出惯性传感器的偏差值,并用来修正无偏差滤波器.因此,应用所提出的滤波方法后,车辆定位系统在不确定噪声的影响下也能获得较好的性能.实验结果显示所提出的交互式多模型两级卡尔曼滤波的平均定位误差比一般两级卡尔曼滤波方法低25%.
In order to solve the problem that it is difficult to obtain good localization performance under the influence of uncertain Kalman filter and two-level Kalman filter on the general state of inertial sensors, a new type of two-level Kalman filter based on interactive multi-model Method to adapt to the uncertain noise of MEMS inertial sensors.According to different noise characteristics, three deviation filters are built to cover a wide range of noise levels.An interactive multi-model algorithm based on three deviation filters can accurately estimate the inertial sensor And used to correct the unbiased filter.Thus, after applying the proposed filtering method, the vehicle location system can obtain better performance under the influence of uncertain noise.Experimental results show that the proposed interactive multi-model two The average position error of Kalman filter is 25% lower than that of two-level Kalman filter.