Font Size: a A A

Design And Key Technologies Research Of Low-cost Indoor Navigation System

Posted on:2023-01-20Degree:MasterType:Thesis
Country:ChinaCandidate:Y GuoFull Text:PDF
GTID:2568306824992119Subject:Civil surveying and mapping and information technology
Abstract/Summary:PDF Full Text Request
After the emergence of GNSS technology,various positioning and navigation services have flourished.However,GNSS cannot locate in indoor environment,resulting in unsatisfactory indoor navigation effect.Therefore,with the increasing demand for location services in indoor environments,indoor precise positioning and navigation services have become a research hotspot.At present,indoor navigation mainly relies on the signals emitted by various sensors pre-arranged in the scene for positioning.The complexity of the indoor scene can easily interfere with the positioning signals,affecting its positioning accuracy and stability.In addition,there is currently a lack of simple and convenient methods for building a full-process navigation system from 3D model generation,indoor stable positioning and real-time navigation.Restricted by these conditions,indoor navigation has not been applied on a large scale.Therefore,improving the accuracy and stability of indoor positioning and reducing the construction cost of indoor navigation system are of great significance to promote the wide application of indoor navigation.In view of the current difficulties in indoor positioning and navigation applications,this thesis designed a low-cost indoor navigation system construction framework.In the indoor precise positioning stage,with the advantages of low cost and stable signals of Bluetooth Low Energy(BLE),this thesis proposed an indoor positioning algorithm that combines the real-time abnormal rates weighting of the signal sources.The signal strength value of the BLE base station is used for abnormal detection,and then a weight matrix is generated according to the abnormal detection results,and finally the coordinates are solved by nonlinear least squares iteration.In the stage of system construction and real-time 3D navigation,this thesis proposed an indoor navigation system construction framework that combines Building Information Modeling(BIM)and Unity 3D.Quickly generate BIM as a 3D map for indoor navigation,and load it into the Unity 3D engine to develop the indoor navigation model after preprocessing.Based on Unity 3D,the main functions of the indoor navigation system such as indoor positioning and real-time 3D navigation are built.The main research contents of this thesis include:(1)Analysis of indoor positioning technology based on BLE.The geometric method and the fingerprinting method are compared and analyzed from the technical principles,advantages and disadvantages of various basic technologies,etc.At the same time,this thesis verified that pedestrians and other environmental factors are prone to interfere with the BLE signal through experiments,which illustrates the necessity of using the geometric method for rapid positioning and overcoming the influence of signal interference.(2)Localization algorithm based on real-time anomaly rates weighting.In this thesis a positioning algorithm based on weighted real-time anomaly rates is proposed.The isolation forest algorithm is introduced to perform real-time anomaly detection on the collected signal strength values,calculate the anomaly rates,normalize the anomaly rates of each signal source,and generate a weight matrix based on the signal strength value.Finally,a damped nonlinear least squares model(Levenberg-Marquadt)is used for iterative coordinate solving.In an experimental scene,the overall positioning accuracy of this method is better than 2m,and it is significantly improved compared with the other three solutions.The improvement ratios are38.69%,36.60%,and 29.52%,respectively,which effectively overcomes the decrease in positioning accuracy caused by the interference of environmental factors on the BLEs.(3)Research on the construction of indoor 3D navigation model.The main methods of indoor space representation and path planning are analyzed.A new indoor 3D navigation model construction method is proposed.The basic process of this method is summarized,the main advantages of this method are analyzed.The basic rules of 3D navigation mesh generation based on Unity 3D are summarized.(4)Indoor navigation system combining BLE,BIM and Unity 3D.This thesis proposed an indoor navigation system construction framework combining BLE,BIM and Unity 3D and completes the development of the navigation system.This framework makes full use of the advantages of simple BIM generation,high accuracy and Unity 3D’s openness and portability,and integrates BLE positioning technology to provide a complete set of low-cost indoor navigation system design idea.
Keywords/Search Tags:Indoor positioning and navigation, BLE, Abnormal detection, BIM, Unity 3D
PDF Full Text Request
Related items