Research On UWB/BLE Integrated Indoor Pedestrian Navigation And Positioning Model

Navigation and positioning is related to the country’s society,economy,security and other aspects.More than 80% of people’s living time is in indoor and underground space.UWB indoor positioning technology had high positioning accuracy but high cost,and BLE indoor positioning technology had low cost but low precision.Indoor positioning space urgently needed a positioning technology and solution with high precision,low cost and easy implementation.Aiming at the problem that a single sensor is difficult to meet the indoor complex navigation and positioning with high accuracy and low cost,this paper carried out the research on the UWB/BLE integrated pedestrian navigation and positioning model,which covered the BLE signal distance model,UWB ranging optimization,BLE positioning algorithm,UWB positioning algorithm,UWB/BLE integrated positioning model and pedestrian navigation constraints.The following results had been achieved:(1)To solve the problem of BLE signal-distance conversion,a multi signal filtering method is proposed based on Gaussian filtering and mean filtering to weaken the strong fluctuation of BLE RSS values,and the range conversion model of segmented BLE signal is constructed based on exponential function and segmented fitting idea in this paper.In the experiment,the fitting accuracy of the segmented BLE signal-distance conversion model is 6% higher than that of the conventional model.In order to improve the ranging quality,the experimental inquiry method is used to analyze the ranging quality of BLE signal transmission angle,BLE protocol version and transmission masking environment.Different influence factors have different effects on ranging accuracy in the experiment.In practical application,the variables should be controlled uniformly.In addition,to cope with NLOS environment,BLE ranging model considering error is established based on peak optimization and threshold correction method,and the ranging accuracy can be improved by 48.99% in human occlusion environment.In addition,BLE wireless rendezvous positioning is realized based on WLS method,and sub-meter precision positioning can be realized in the experiment.(2)Aiming at the systematic deviation of UWB ranging values,this paper uses the sample database and BP neural network to construct the functional relationship between the original UWB ranging value and the real distance,and the UWB ranging optimization model is constructed based on the functional relationship and systematic error correction technology to eliminate the systematic deviation error in the UWB ranging value.Under the experimental environment and equipment conditions,the maximum ranging error within 25 m after optimization of UWB ranging optimization model is reduced by 80.0%,the average value is reduced by 76.9%,and the accuracy reliability less than 5cm is 87.7%.In the simulation comparison experiment,the positioning experiment is carried out after UWB ranging optimization.The systematic offset of the positioning trajectory is obviously eliminated,the maximum positioning error is reduced from 0.5m to 0.12 m,and the positioning accuracy is significantly improved.(3)Aiming at the contradiction between positioning accuracy and positioning cost,a UWB/BLE integrated RKF positioning algorithm is constructed based on UWB and BLE,and a pedestrian navigation constraint algorithm is designed based on speed constraints and road network constraints.According to the positioning accuracy requirements,the UWB/BLE pedestrian navigation and positioning scheme is designed.The indoor pedestrian navigation and positioning area is divided into high-precision positioning areas,common positioning areas and intermediate transition areas.Different areas call different positioning algorithms to achieve combined positioning.In the real experimental scenario,in the UWB positioning area,the BP neural network ranging optimization can significantly improve the positioning quality,and the UWB positioning algorithm can achieve high-precision positioning better than 0.1m,with the positioning accuracy improved by 70.64%.In BLE positioning area,the positioning accuracy can reach sub meter level.In the UWB/BLE combined positioning area,the accuracy of traditional uwb/ble combined positioning is higher than that of ble positioning area and lower than that of UWB positioning area.The positioning accuracy of UWB/BLE integrated positioning considering pedestrian navigation constraints is significantly higher than that of traditional uwb/ble integrated positioning,and the positioning trajectory is closer to the real path.