基于CNN和在线学习的UWB室内定位算法

UWB Indoor Positioning Algorithm Based on CNN and Online Learning

为了减小非视距(NLOS)误差对超宽带(UWB)室内定位系统定位精度的影响,提出了一种基于卷积神经网络(CNN)的超宽带室内定位算法。利用超宽带系统采集非视距环境下的室内定位数据,根据信号在非视距环境下传播时的误差特性建立CNN模型,将定位数据输入网络进行训练,以减小NLOS误差对定位精度的影响,然后用扩展卡尔曼滤波(EKF)进行位置估计;当系统处于不同室内环境时,使用在线学习算法调整CNN参数,提高系统的兼容性。实验结果表明,该算法可以在不同室内环境下有效减小NLOS误差的影响,保持厘米级的定位精度,具有一定的实用价值。

In order to eliminate the effects of non-line-of-sight(NLOS)error on the positioning accuracy of ultra-wideband(UWB)indoor positioning system,an UWB indoor positioning algorithm based on convolutional neural network(CNN)is proposed.By using UWB system to collect the indoor positioning data and according to the error characteristics of signal propagation in the NLOS environment,a CNN model is established and the data is input into the network for training,so as to reduce the influence of NLOS error,and the position was estimated by extended kalman filter(EKF).When the system is in different indoor environments,the online learning algorithm is used to adjust the CNN parameters to improve the system compatibility.Experimental results show that this algorithm can effectively reduce the influence of NLOS error in different indoor environments,and maintain the positioning accuracy of centimeter level,which has certain practical value.