Research On Positioning In Complex Indoor Environment Based On UWB And IMU

Ultra-wide Band(UWB)positioning technology has become one of the most promising indoor positioning technologies due to its high accuracy,good delay resolution,low power consumption and strong robustness in complex indoor environments.However,due to the complexity of indoor space,the non-line-of-sight state(NLOS)caused by various obstructions will cause serious positioning deviations in UWB positioning.NLOS in a complex indoor environment is an important factor affecting the development of UWB indoor positioning.Therefore,this dissertation studies the NLOS recognition and positioning error mitigation of UWB positioning in a complex indoor environment and proposes an effective solution.NLOS recognition is a prerequisite for mitigating UWB positioning errors,and it is critical to the accuracy of UWB indoor positioning.Existing research only recognizes the line-of-sight state(LOS)and NLOS and ignores the contribution of the occlusion category to the perception of spatial information.Therefore,this dissertation proposes a two-way search algorithm(BS-mRMRMC)based on maximum correlation,minimum redundancy and minimum calculation cost.By setting the maximum evaluation index constraint threshold and the calculation cost constraint threshold,the optimal channel impulse response can be determined feature set.At the same time,based on the vector projection method,a hierarchical structure of Decision Tree Support Vector Machine(DT-SVM)is designed to verify the accuracy of recognition of each category.The feature set determined based on BS-mRMRMC can accurately identify LOS and NLOS occlusion categories.Experiments show that based on the same number of UWB CIR signal features,the algorithm has an average recognition accuracy of 95.7%for each occlusion category,which is better than the other three.A feature selection method and occlusion category can provide more effective information for UWB indoor spatial information perception.On the basis of NLOS recognition,on the basis of using only three UWB positioning base stations,two positioning subsystems,UWB and inertial navigation,are used for data fusion.UWB's high-precision positioning under LOS,and IMU positioning are not affected by the external environment.This dissertation designs a UWB positioning error mitigation system in complex indoor environments based on the tightly coupled data fusion of UWB and IMU,and using the Unscented Kalman Filter algorithm for direct filtering.The results of the final experiment prove that the UWB positioning error mitigation system designed in this dissertation can effectively alleviate the positioning error caused by NLOS in the complex indoor environment,whether it is under less occlusion or more occlusion.Especially when there is a serious NLOS impact in a complex indoor environment,high-precision positioning can still be achieved when only three base stations are used for UWB positioning,and the root mean square error is only 11cm.