| In recent years,the research on indoor high-precision navigation and positioning technology has highlighted great value.In many research fields,ultra-wideband(UWB)and inertial navigation system(INS)technology has become a research hotspot because of its irreplaceable advantages.In this context,this paper carried out research on UWB positioning technology and pedestrian inertial navigation positioning technology,and also carried out research on the field of multi-sensor combined positioning.The main research contents are as follows:The paper first introduces the theory of UWB positioning technology,the theory of inertial navigation solution and the algorithm of multi-sensor data fusion.In order to deeply study the characteristics and principles of UWB positioning technology and the performance of UWB positioning technology in different ranging positioning methods,theoretical research is carried out on UWB ranging methods,time-of-arrival(TOA)positioning algorithms and time-difference-of-arrival(TDOA)algorithms.Localization experiments are carried out in real environment.The positioning performance of TOA positioning and traditional TDOA positioning is compared and studied through static single point positioning experiments and dynamic tracking positioning experiments.The actual analysis shows that the TOA positioning algorithm based on bilateral two-way ranging(DS-TWR)performs better than the TDOA algorithm.In order to solve the problem of error accumulation in pedestrian inertial navigation and positioning,this paper proposes a multi-condition zero-speed detection scheme based on the characteristics of inertial sensor data,and optimizes the judgment performance of the zero-speed detection module.At the same time,the error correction method of ZUPT+ZARU+Attitude self-observation is proposed.In addition to the effective correction of velocity and angular velocity errors by zero-speed correction(ZUPT)and zero angular velocity correction(ZARU),this method also has the ability to correct certain attitude angle errors.A foot-mounted indoor pedestrian navigation and positioning system based on ZUPT+ZARU+Attitude self-observation and correction method was built.The results of comparative experiments show that the performance of the proposed optimization and improvement scheme is improved by 35.15% compared with the EKF-ZUPT algorithm,and 15.27% compared with the performance of EKF-ZUPT+ZARU,which verifies the effectiveness of the proposed improved scheme algorithm.In order to solve the problem of poor positioning accuracy of UWB in NLOS environment.Considering the respective advantages and disadvantages of UWB and INS methods,and giving full play to the advantages of UWB high precision and INS not affected by the environment,a UWB/INS tight combination positioning system based on Rank Filter(RF)is proposed.In this system,a rank filtering algorithm with better filtering effect than EKF and UKF is adopted.Through experiments in the NLOS environment,the positioning errors of the combined system are obtained: the maximum error(1.640m),the minimum error(0.115m),and the average error(0.723m)are all the smallest.Compared with the single positioning technology,the positioning accuracy of this scheme has been significantly improved,which verifies the effectiveness of the tight combination system. |
