Research On Indoor Positioning Technology Based On Multi Feature Fusion

With the promotion of mobile smart device technology,people's demand for Location Based Services is becoming more and more common.Meanwhile people have higher requirements for the accuracy of location services.Although the traditional positioning system can be well applied in the open outdoor environment with less serious occlusion,it does not perform well in the indoor environment with serious occlusion.Due to the needs of indoor work and life,the demand for indoor positioning service will be increasing in the near future.In indoor environment,GPS signals can not penetrate the building,and the propagation of wireless signals may be affected by reflection,reactance and multipath effect.Due to these factors,positioning in indoor scene is far more difficult than outdoor positioning.Starting from the indoor image positioning and UWB positioning,the image positioning database is constructed,and the improved image feature fusion model proposed is applied to design and implement the indoor image positioning system based on image.For UWB positioning,a set of indoor positioning system based on UWB is designed,the specific software design flow and hardware schematic diagram are given,and stable and accurate UWB positioning is realized.On this basis,the video car is selected as the experimental carrier,and the indoor positioning system based on the fusion of image and UWB is finally designed and realized,and the high-precision indoor positioning is completed.The main work and innovation points are as follows:Firstly,the principle of UWB positioning is analyzed.The data acquisition scheme used is determined by combining the different acquisition methods of UWB ranging and the corresponding subsequent processing methods.In view of the unstable situation of UWB ranging,an improved ranging algorithm is proposed to reduce the ranging fluctuation error.Aiming at the problem of UWB positioning coordinate value mutation,the improved Kalman filter is used to optimize it.And a set of indoor positioning system based on UWB is designed and developed,the hardware schematic design and software development process are given.Furthermore,the ranging and positioning accuracy are improved through error calibration.Finally,the reliability and accuracy of the UWB positioning system designed are tested and verified in the actual environment accuracy.Secondly,the image preprocessing,image feature extraction and image classification algorithms involved in image location are introduced.Aiming at the problem of insufficient image description by single feature,an improved image feature fusion algorithm model is proposed,and the specific design idea and structure model block diagram are given.The effectiveness and accuracy of the model are verified by simulation and actual test.Finally,based on the improved algorithm model,the image positioning system including the computer interface software is designed.It runs well in the actual test.Thirdly,according to the inaccurate positioning of UWB in the signal unstable environment,a fusion indoor positioning system based on image and UWB is designed.By using the video car as the experimental carrier and the car camera as the data input interface of image positioning,the UWB label node is installed on the car to achieve UWB positioning,and the dynamic weight mechanism is designed to effectively combine the image positioning coordinates and the UWB positioning coordinates.Through the actual positioning test,the accuracy of the fusion positioning algorithm designed in this paper is verified.The positioning system that can be applied to indoor environment is built,which can meet the needs of high-precision indoor positioning system.