Font Size: a A A

Design And Implementation Of Indoor Positioning Tracking System Based On UKF With Fusion Of WiFi And Dead Reckoning

Posted on:2019-11-07Degree:MasterType:Thesis
Country:ChinaCandidate:D K ZhaFull Text:PDF
GTID:2428330590475446Subject:Software engineering
Abstract/Summary:PDF Full Text Request
With the development of science and technology and the improvement of people's living standard,indoor positioning and navigation technology becomes one of the most important technologies used in people's life.The wireless location technology and the dead reckoning are research hotspots of indoor positioning.Wireless positioning technology can realize localization with low cost.However,its positioning accuracy and reliability can't meet demand.Dead reckoning can lead to high precision navigation with an initial position,but the error increases with time.Thus,people's demand of indoor positioning cannot be met.After analyzing the fusion technology of indoor positioning,the indoor positioning based on the unscented kalman filter(UKF)algorithm with fusion of WiFi fingerprint positioning and dead reckoning is realized in the thesis.Firstly,a fusion positioning model based on unscented kalman filter is established.Predict model,measurement model and fusion model parameters are analyzed for model error.According to the results of the analysis,WiFi fingerprint positioning and dead reckoning are optimized.Kalman filter,fingerprint database with direction and multi-threshold peak value gait detection are used to solve problems like the signal fluctuation,personal shade and error of heading estimation etc.Then the parameters of the fusion model are adjusted according to the error of WiFi fingerprint positioning and dead reckoning.The positioning system based on client and server is implemented,which improves the usability and extensibility.In the experiment,over 90% of the points are within 2 meters from true points and cumulative error in a circle is within 1.3 meters;average positioning period is within 1.3 seconds,which meets the design requirements.The indoor positioning based on the unscented kalman filter algorithm with fusion of WiFi fingerprint positioning and dead reckoning in the thesis meets the design demand and performance indicators,which is of value for indoor positioning system in engineering.
Keywords/Search Tags:unscented kalman filter, indoor positioning, fusion algorithm, dead reckoning, fingerprint location
PDF Full Text Request
Related items