UWB/INS Based Indoor Positioning and NLOS Detection Algorithm for Firefighters

Abstract

In order to detect and mitigate the NLOS (non-line-of-sight) error of UWB (ultra-wideband) in complex fire scene environment to obtain the accurate positions of firefighters, an indoor positioning algorithm integrating UWB and INS (inertial navigation system) is proposed. The NLOS detection of UWB and the data fusion of UWB/INS are completed under the designed two-stage EKF (extended Kalman filter) framework. Firstly, the first-stage EKF is used for INS to obtain the initial estimated position of firefighter, and then the position is converted into the distance value to each base station and compared with the current UWB ranging values to detect the NLOS errors. Then, the residual matrix is calculated according to the detection results to dynamically adjust the measurement noise matrix of the second-stage EKF, so as to mitigate the NLOS errors and obtain the joint position estimation of UWB/INS. Finally, the joint positioning result of the second-stage EKF is used as the feedback value of the whole system for the next positioning. In order to verify the performance of the proposed algorithm under different degrees of NLOS error, simulation experiments and field experiments are designed. Compared with the state-of-the-art algorithms mentioned in the paper, the proposed NLOS detection algorithm has good detection ability and strong stability, and UWB/INS fusion mode can effectively improve the positioning accuracy.