Research On Pedestrian Integrated Navigation And Positioning Technology Based On GNSS/IMU

With the increasing demand for indoor navigation and positioning services,the integrated navigation and positioning technology based on GNSS and MEMS IMU has become a research hotspot in the field of indoor and outdoor seamless positioning because of its complementary advantages and strong autonomy.However,the cost or accuracy of MEMS IMU for pedestrian positioning in the market limits its promotion and development.In this paper,the characteristics of low-cost ins are studied.By analyzing the characteristics of pedestrian motion,the zero speed correction and heading correction models for sins algorithm are proposed,which effectively limit the rapid divergence of IMU error.Considering the emergency positioning and other scenes and people's demand for indoor and outdoor seamless positioning accuracy,a loose combination model based on GNSS single point positioning and IMU is established,which improves the outdoor positioning accuracy of pedestrians and ensures that pedestrians can still obtain absolute position information indoors.The main work of this paper is as follows(1)According to the gait feature extraction of pedestrians in different motion states,the supporting phase and swinging phase can be divided into two phases,which provide the necessary prerequisite for zero speed detection.Since the traditional single threshold zero speed detection algorithm is prone to misjudgment and missed judgment,a generalized likelihood ratio detection method is proposed to calculate the maximum likelihood estimation of angular velocity and acceleration through probability,and then judge the support phase of the traveler according to the generalized likelihood function.The appropriate sliding window is selected to make statistics of zero speed detection under different motion states.The results show that GLRT has the best detection effect,and the error rate is controlled within 3%.(2)Aiming at the problem of position and heading error divergence of low-cost IMU,a method of zero velocity correction(ZUPT)and zero integral heading rate correction(ZIHR)is proposed.The system error is estimated by using the velocity information of MIMU output as observation,and the navigation information is corrected by using the estimated value of system state.With the help of ZUPT idea,the difference of heading angle between the front and back sampling points is taken as the measurement quantity to suppress the heading error angle.The experimental results show that the positioning accuracy of pedestrian sins algorithm based on ZUPT + ZIHR in short distance is decimeter level,which fully meets the needs of pedestrian indoor positioning service.(3)Aiming at the scene of pedestrian indoor and outdoor seamless positioning,by considering the cost of hardware system and people's requirements for positioning accuracy,through the analysis of inertial navigation error equation,the state equation of 15 dimensional state variables is established,and the loose combination fusion algorithm is designed by using GNSS pseudo range single point positioning results and INS output position and speed error as observation,the positioning accuracy is about 30% higher than that of single point positioning.In addition,GNSS has high accuracy in short-term velocity measurement.It calculates the velocity and calculates the relatively accurate course by using the Doppler original observation value,and gives the initial absolute position and course to the inertial navigation system,so as to ensure that the absolute position information can still be obtained in the indoor environment.