研究是建立在基于农业物联网技术的农业大棚信息监测系统的基础之上,利用卡尔曼滤波算法对采集终端传感器采集到的大量数据进行融合处理.该算法首先通过状态方程和节点观测方程推出状态一步预测值方程,其次将状态一步预测值方程与节点观测方程结合,便可得到系统状态最优估计值.最后通过预测值与测量值进行比较,并结合协方差的变化对状态估计值进行加权修正.仿真结果分析表明,利用卡尔曼滤波算法进行数据融合,能够得到更加平稳的数据值,精度也大大提高.
This study is basic of the agricultural information technology monitoring system based on agricultural Internet of Things technology. It was used Kalman filtering algorithm to fuse mass data collected by acquisition terminal sensors. Firstly, the state one-step predictor equation is derived by the state equation and the node observation equation. Secondly, the state one-step predictor equation is combined with the nodal observation equation to obtain the optimal estimation of the system state. Finally, the predicted value is compared with the measured value, and the state estimation value is weighted and modified in conjunction with the change of the covariance. The simulation results show that using Kalman filter algorithm for data fusion can obtain more stable data values, and the accuracy is greatly improved.