基于BP神经网络补偿卡尔曼滤波的UWB精定位算法

Application of UWB precise positioning based on Kalman filter for power field operation safety

本文研究典型应用场景下电力现场作业的超宽带精确定位,提出一种基于BP神经网络补偿卡尔曼滤波的UWB精定位算法。首先在某段时间内利用定位系统测得全部采样时刻的初始观测解,得到初始观测解集;其次根据卡尔曼滤波器的状态方程,初始观测解集输入卡尔曼滤波器;最后用BP神经网络补偿卡尔曼滤波,在最小均方误差下求得定位系统的二维状态向量的估计值。通过仿真表明:该补偿算法在原卡尔曼滤波算法上提升了定位精度。

In this paper,we study the ultra-wideband precise positioning of power field operation under typical application scenarios,and propose a UWB fine positioning algorithm based on BP neural network compensation Kalman filter.Firstly,the initial observation solution of all sampling moments is measured by the positioning system in a certain period of time to obtain the initial observation solution set.Secondly,according to the state equation of the Kalman filter,the initial observation solution is input to the Kalman filter;finally,the BP neural network is used.The compensated Kalman filter is used to obtain an estimate of the two-dimensional state vector of the positioning system under the minimum mean square error.The simulation shows that the compensation algorithm improves the positioning accuracy on the original Kalman filter algorithm.