Research On Multi-source Heterogeneous Personnel Positioning And Navigation System Based On Android Platform

With the rapid development of modern information technology,smart phone terminals,wearable devices and other technologies are becoming more and more mature,and LBS is loved by more people.According to surveys,80%-90% of people's activities happen indoors,and the demand for LBS in indoor environments is increasing day by day.However,due to the limitation of positioning time,positioning accuracy,and complex indoor environment,the actual availability of a single indoor positioning technology is not ideal.Therefore,how to combine multiple positioning technologies to meet indoor personnel positioning needs has become a current research hotspot.Based on this,the thesis mainly uses inertial positioning technology,Wi Fi positioning and geomagnetic matching technology as the assistance to carry out research on multi-source heterogeneous personnel positioning and navigation system based on android platform.The main research contents are as follows:1.For inertial positioning,MEMS sensor error compensation algorithm is designed,and a weighted fusion algorithm is used to correct the cumulative heading error.Aiming at Wi Fi positioning,a KNN Wi Fi fingerprint matching algorithm based on K-means clustering is proposed,which can improve the system accuracy and reduce the calculation amount of positioning.For geomagnetic positioning,DTW algorithm is introduced for geomagnetic fingerprinting matching,even if the geomagnetic fingerprint database waveform and the number of geomagnetic waveforms to be matched are different,the matching can be performed well.2.Taking into account the high accuracy but unstable problem of Wi Fi positioning,the complementary features of low geomagnetic positioning accuracy but high stability,a combined positioning method that uses Wi Fi positioning results to constrain the range of geomagnetic matching is proposed.Experiments show that the combined positioning accuracy is about 24% higher than Wi Fi positioning,about 13% higher than geomagnetic positioning,trajectory reduction is 9.22% higher than Wi Fi positioning,and 4.35% higher than geomagnetic positioning,which effectively improve the situation of Wi Fi positioning fluctuation and geomagnetic positioning mismatch.3.In order to fuse combined positioning results and inertial positioning results,a decision-level fusion positioning algorithm based on genetic algorithm to optimize BP neural network is proposed,and implemented in the smart phone terminal.Experiments show that the accuracy of fusion positioning is improved by 69%,66%,61%,and 56%compared with inertial positioning,Wi Fi positioning,geomagnetic positioning,and combined positioning;the average trajectory reduction degree is increased by 39.33%,45.50%,40.63%,36.28%.It further shows that the algorithm used for multi-source information fusion positioning can significantly improve the positioning accuracy of indoor personnel,and has a better positioning effect.