Research On Indoor Target 3D Tracking And Positioning Technology Based On UWB

With the rapid development of economy and technology,people have an increasing demand for indoor location-aware services.Among them,high-precision,low-latency,and non-line-of-sight(NLOS)indoor tracking and positioning technology has become a research focus,and it has attracted more and more attention from experts,scholars and related industries.Compared with other indoor tracking and positioning technologies,the ultra-wideband(UWB)tracking and positioning technology has the advantages of fast transmission rate,low power consumption,and high accuracy.It can better solve the problem of indoor target three-dimensional tracking and positioning.This paper conducts in-depth research on ultra-wideband ranging and tracking and positioning technology.The main research contents and achievements are as follows:First of all,this paper has carried out theoretical research on ultra-wideband technology and related positioning technology.The common positioning information acquisition schemes are compared and analyzed from the aspects of difficulty and accuracy,and a high-precision,easy-to-implement double-sided two-way ranging scheme(DS-TWR)is selected.The error factors in the UWB tracking and positioning system are analyzed,and the distance correction method based on least squares is proposed for the antenna delay problem;and the weighted mobile ranging filter method based on exponential smoothing is proposed for the positioning delay problem.In order to evaluate the performance of the tracking and positioning system,the evaluation indicators of the tracking and positioning system such as system capacity,positioning accuracy and tracking delay are proposed,and they are used in simulation experiments and indoor tracking and positioning experiments.Then,this article uses the STM32 microprocessor,equipped with the self-developed ultra-wideband receiving and transmitting module,to design and implement the ultra-wideband double-sided two-way ranging scheme.According to the double-sided two-way ranging process,the UWB ranging communication message format,the single-tag ranging process and the multi-tag ranging scheme are designed,and the event-driven finite state machine is used to implement the double-sided two-way ranging process between the tag and the base station.Experimental tests have verified that the UWB double-sided two-way ranging solution can achieve high-precision and stable indoor ranging.Next,in order to realize the tracking and positioning of indoor moving targets,this paper conducts in-depth research on the ultra-wideband 3D positioning algorithm.Aiming at high-precision and low-latency tracking and positioning requirements,the Chan positioning algorithm,Taylor positioning algorithm,and extended Kalman filter tracking and positioning algorithm tracking and positioning performance are analyzed and compared,and the Chan-Taylor-Kalman joint three-dimensional tracking and positioning algorithm based on residual discrimination is proposed.Location Algorithm.The C-T-K joint tracking and positioning algorithm uses the NLOS residual discrimination method to perform NLOS identification on the initial estimated position obtained by the Chan algorithm to identify whether there is a NLOS error in the data.In the case of NLOS error,the Taylor positioning algorithm is used to iterate the initial estimated position to suppress the NLOS;for the LOS situation,a single Chan positioning algorithm is still used for calculation to reduce positioning delay.Finally,the Kalman filter algorithm is used to reduce the jitter of the positioning result and improve the tracking and positioning effect of the algorithm.Through simulation experiments,the tracking and positioning performance of the four positioning algorithms are compared and analyzed,and the effectiveness of the C-T-K joint tracking and positioning algorithm is verified,and it has a certain inhibitory effect on NLOS errors.In order to further verify the performance of the tracking and positioning algorithm,this paper designs an UWB 3D tracking and positioning system,which uses spotlights to track the position of moving targets.Experiments have proved that for fixed targets,the CTK joint tracking and positioning algorithm improves the stability of positioning and achieves high-precision indoor positioning for fixed targets;for moving targets,the CTK joint tracking and positioning algorithm achieves low latency in the LOS situation Tracking and positioning,the spotlight can follow the target in real time.Finally,in order to further solve the impact of NLOS errors,this paper conducts a research on deep learning positioning algorithms.According to the time continuity of UWB ranging,anti-NLOS location algorithm based on LSTM network model is proposed.The NLOS simulation experiment proves that the LSTM network model positioning algorithm has a good suppression of the NLOS error and improves the positioning accuracy.