Study On Data Fusion Method Of Distributed INS/UWB Indoor Integrated Positioning

With the rapid development of science&technology and society,there is a high demand for the accuracy,reliability and anti-interference of positioning systems in some complex environments.Although Global Navigation Satellite System(GNSS)has been widely used in fields of moving vehicles such as automobiles,ships and aircraft,the fact that GNSS signals are easily blocked in complex environments such as tunnels,interior of high buildings or enclosed factory areas,results in a remarkable decrease in positioning accuracy.In order to achieve the high-precision of indoor location and its accuracy,stability and versatility,indoor positioning has gradually been a hotspot for research.To improve the accuracy of indoor positioning,this paper studies the integrated information fusing method of Inertial Navigation System(INS)/Ultra Wide Band(UWB)combined system for complex indoor environments.Integrated information these two sensors was carried out to be fused in order to precisely locate the target pedestrians in complex indoor environments.The detailed contents are as follows:Firstly,targeting complex indoor environments,the pros and cons of technologies of WiFi,radio frequency identification,ultrasonic,UWB,and INS positioning are studied.Among them,the UWB positioning system has the pros of strong penetration,high transmission speed and low transmission power,but its positioning accuracy is lack of uncertainty due to the varying density of prearranged base stations,that is to say,the denser base stations are,the higher the equipment cost is.The pros of INS can be also easily seen--strong stability,strong anti-interference ability and high short-term accuracy whereas the location errors of the calculation are accumulated to a large amount over time.Combining the pros and cons of the two sensors,the INS/UWB combined-sensors pedestrian positioning model and the experiment platform are both built up.The positioning model includes INS positioning system,UWB positioning system and data fusion filter.The INS positioning system and the UWB positioning system operate in parallel.The data fusion filter is composed of a plurality of local filters operating in parallel and a main filter.Furthermore,a dual-distributed Extended Kalman Filter(EKF)combined positioning algorithm is proposed for the purpose of reducing the influence of system parameters on the positioning accuracy.Each local filter contains two sub-filters: a distributed EKF state vector filter and a distributed EKF system parameter filter.The distributed EKF state vector filter estimates the location and velocity error while the distributed EKF system parameter filter estimates the state matrix.The observation vector is the difference between the distances measured by INS and UWB.The main filter integrated the estimated value of the local filter to estimate the location error of the target pedestrian at one moment and ultimately the optimal value of position information is obtained.Additionally,to reduce the influence of Colored Measurement Noise(CMN)on the positioning accuracy,a distributed Kalman Filter(KF)algorithm for colored measurement noise,that caculates to pick out the optimal CFM in virtue of a variable Colored Factor Matrix(CFM)which predicts a more accurate position information of the target pedestrian at one moment by Mahalanobis distance,is proposed in this study.Finally,to ensure the positioning accuracy and the system stability,the state vector introduces position information such as attitude angle,gyroscope bias,and accelerometer bias,which can ensure the positioning accuracy and the collected data can be used more efficiently but it causes an increase the complexity of computation.Thus,to maintain the pros of nonlinear filtering methods in dealing with nonlinear problems and reduce the computational complexity,a distributed Cubature Kalman Filter(CKF)/KF hybrid filtering algorithm is proposed,which ensures the balance between reducing calculation time and maintaining high positioning accuracy of CKF.In order to verify the effectiveness and reliability of the system,experimental tests are carried out in various scenarios to reveal the result indicating that the accuracy of integrated fusing algorithm can reach the centimeter level,which is better than that of UWB or INS single sensor.The interference of positioning accuracy caused by the uncertainty of system parameters and the uncertainty of colored measurement noise is effectively suppressed.It also reduces computation time and improves the positioning accuracy and reliability of the INS/UWB combined positioning system.