Research On Pedestrian Autonomous Navigation Based On Multi Source Information Fusion In Forest Region

In the forest area where GNSS signal is not available or missing,the inertial pedestrian positioning system based on micro inertial measurement unit(MIMU)has become an important choice for forest operators(firefighters,inspectors,etc.).In view of the characteristics of high canopy density and easily blocked satellite signals in forest area,the autonomous attitude determination and positioning of forest operators are set as the research object,and the platform of pedestrian autonomous positioning system is designed and built.The main research contents are as follows:Zero Velocity Update(ZUPT)is an effective solution to reduce the cumulative positioning error caused by the integral calculation of the inertial navigation.The accuracy of conventional ZUPT for attitude alignment is reduced by the zero offset of acceleration and the drift of gyroscope during the standing phase.An initial alignment algorithm based on Adaptive Gradient Descent Algorithm(AGDA)is proposed.At the stepping phase,the measurement noise of Extended Kalman Filter(EKF)filter is affected by the measured acceleration and angular velocity of the sensor under high frequency dynamic condition.A Double-constrained Extended Kalman Filtering(DEKF)is proposed.An adaptive parameter positively correlated with the acceleration and angular velocity is set,and the measurement noise in the DEKF is adaptively adjusted.The low accuracy and high noise of MIMU have a serious impact on the measurement accuracy of zero velocity detector,which will lead to poor 3D pedestrian positioning accuracy.An improved generalized likelihood ratio test(GLRT)method is proposed.The core is to dynamically adjust the detection threshold according to different motion states,so as to determine the constraint relation for detection.In addition,the drift of sensors and the accumulated error of position integration affect the accuracy of 3D attitude and position.Therefore,an improved extended Kalman particle filter(EKPF)is proposed.The improved EKPF constructs the position constraint model in altitude direction by introducing barometer information.The new particle filter distribution function is constructed by constrained multi-source observation information,which makes the posterior probability distribution closer to the real distribution.Magnetic field measurement is easily affected by external magnetic interference,which leads to heading error in the forest area with magnetic interference such as ore and metal.A magnetic interference test method based on generalized likelihood ratio test(GLRT)is proposed in this study.A likelihood ratio function is constructed to maximize the phase probability of magnetic interference,so that the inequality relation for detection can be determined.In addition,the combination of ZUPT with EKF and EKPF can suppress accumulated errors,but the heading deviation cannot be accurately estimated.Therefore,an improved EKPF is designed to suppress the magnetic interference,and the heading observation is compensated by adding adaptive parameters.In view of the serious defects of jitter and random noise of single inertial sensor on the soft ground of forest area,a two-point zero position difference ZUPT combined with foot and lower leg kinematics information was proposed in this work.Firstly,the zero position difference constraint matrix with ankle joint as the connection point is established by analyzing the local motion characteristics of foot and lower leg.In order to get accurate attitude and position by fusing the information of foot and lower leg,an improved EKPF based on the zero position difference between the instep and the lower leg is proposed.Moreover,according to the gait information characteristics of instep and lower leg,a weight assignment strategy is designed to detect the zero velocity point of pedestrians.Finally,the experimental platform is built in different scenarios,and the feasibility and effectiveness of the proposed improved pedestrian autonomous positioning system in forest area are verified.The experimental results demonstrate that AGDA method and DEKF method can significantly reduce the attitude and position errors of the positioning system.In the 3D pedestrian autonomous positioning system,the improved GLRT method performs well in detecting the zero velocity point of the foot,and the improved EKPF method can effectively reduce the 3D positioning error.In the pedestrian positioning system with magnetic interference,the false detection rate of the improved magnetic interference detector is significantly reduced,and the positioning accuracy of the improved EKPF based on adaptive heading repair is significantly improved.The ZUPT-Aid method based on two-point zero position difference can significantly enhance the performance of pedestrian autonomous positioning system.