基于AKF-AM-Bi-GRU优化的UWB室内定位方法OA

Aiming at the problem of low accuracy of UWB positioning technology in indoor complex environment,a UWB indoor positioning method based on AKF-AM-Bi-GRU optimization is proposed.Firstly,adaptive Kalman filter based on residual adjustment is used to reduce the influence of signal noise and outliers on UWB plositioning accuracy.Secondly,AM Bi GRU is used to deeply mine the time series characteristics in the historical distance value data to optimize the distance value from the tag point to each base station.Then,the weight of the least squares method is adjusted in combination with the standard deviation of the historical distance value,and the coordinates of the label points are calculated using the weighted least squares method.Finally,it is verified by setting up several groups of real indoor scene experiments,and summarizes the positioning error.The experimental results show that the positioning accuracy of UWB is improved by 44.45％on average after optimized UWB positioning by adaptive Kalman filter based on residual adjustment;Compared with AKF and AKF-AM-Bi-LSTM,the positioning accuracy of AKF-AM-Bi-GRU has increased by 59.08％and 16.35％on average,and the positioning error has reached 5.051 6 cm and 10.034 6 cm in sight distance and non sight distance environments.The proposed method effectively improves the accuracy and robustness of UWB positioning in indoor complex environment,and achieves better positioning results.