Research And Implementation Of UWB-based Indoor Location Algorithm In Mixed Environment Of LOS And NLOS

In recent years,there has been an increasing demand for high-precision location information,and indoor positioning technology has gradually become an important part of smart cities and smart life.The many advantages of ultra wide band(UWB)technology enable this technology to achieve high-precision indoor positioning.Non-line-of-sight(NLOS)errors in the indoor environment are the most important factors affecting the positioning accuracy.This thesis studies the positioning algorithm to suppress NLOS errors,and builds a UWB-based positioning system platform to provide high-precision positioning solutions in the actual mixed LOS and NLOS environment.Firstly,this thesis uses the double-sided two-way ranging(DS-TWR)algorithm of three-time communication to realize the ranging of both sides,and the best matching function is used to optimize the ranging results.Aiming at the errors produced by the traditional differential barometric altimetry in practical applications,an enhanced differential barometric altimetry is proposed.Based on the data of distance measurement and height measurement,a differential UWB three-dimensional positioning algorithm is proposed.The optimal reference base station is selected by NLOS signal identification,and the position coordinates are solved by using the rotation and translation of the coordinate system.Experimental test results indicate that the average positioning accuracy of this algorithm is about 18 cm in the general mixed LOS and NLOS environment;if in the LOS environment,the average positioning accuracy is about 13 cm.Secondly,to solve the problem of low positioning accuracy in more complex positioning scenarios,this thesis proposes a joint positioning algorithm based on Gauss-Newton iteration..The algorithm uses the sub-optimal positioning results of multiple targets to construct a nonlinear model of the positioning error,introduces the contribution factor(33),and obtains the initial estimated coordinates of the target through multiple iterations;Combined with cubature kalman filter(CKF)theory,the coordinates of the first estimate are estimated twice.The analysis of the experimental results shows that the positioning performance of the joint positioning algorithm is superior to the trilateration positioning algorithm and the differential UWB positioning algorithm in this scenario.In static positioning,the average positioning error is 18.4cm,which is 56.7% higher than the trilateration positioning algorithm,and 16.7% higher than the differential UWB algorithm;in dynamic positioning,it has better smoothness and stability than the trilateration positioning algorithm and differential UWB positioning algorithm,and is closer to the actual moving trajectory,which is applicable to most positioning scenarios.Finally,this thesis designs and builds the software and hardware platform of UWB real-time positioning system.The hardware platform components are briefly described,and the hardware design block diagram is given.The implementation process of the underlying software and upper computer software are elaborated in detail.The positioning algorithms proposed in this thesis are all implemented on this platform,and the system has high robustness and good real-time performance.