An Adaptive Pedestrian Navigation Method Focusing On Complex Circumstances

The Individual-soldier navigation system based on information collected by multi-sensor can realize autonomous navigation in complex environments,providing all-weather,accurate and rich positioning and navigation information.It can be used in anti-terrorist activities,fire-fighting,emergency rescue,lumbering,and modern wars,to name but a few.Therefore,the individual-soldier navigation technology based on information collected by multi-sensor is a popular research orientation with bright future.In this essay,self-adaptive individual-soldier navigation system in complex environments based on information collected from MIMU,magnetometer and visual sensor is studied.The primary research contents are as follows:(1)The further comprehensive study on strapdown inertial navigation algorithm has been done,optimizing the the algorithm of Kalman filter based on system errors and Zero-Velocity Update(ZUPT)methods.The installation scheme for MIMU is designed as well as analyzed.The positioning accurate of the test results by individual-soldier navigation strapdown inertial navigation algorithm are analyzed.Based on this,the self-adaptive individual-soldier navigation system scheme including inertial module,ZUPT module,drift calibration module and visual/INS module is designed.(2)The research on three-axis magnetometer calibration has been done.To overcome the unobservability in SINS,a drift reduction method based on geomagnetic information is designed.By judging whether the geomagnetic information is correct,the interference caused by human motion can be reduced.In addition,A self-adaptive navigation based on the stability of the geomagnetic field is studied,which results in a better positioning accuracy.Simultaneously,the stability of the system is strengthened.(3)The unified temporal and spatial calibration for visual/inertial is studied.Through the calibration process,the affective factors of the camera calibration as well as unified calibration of camera and MIMU are summarized.The experiment scheme is designed to testify the accuracy of the calibration results when there is no standard dataset result.(4)Do research on ORB-SLAM.By the theory of preintegrated,the visual/inertial SLAM based on preintegrated IMU is designed,which improves the efficiency of the algorithm.Through the slight slope experiment,it can be testified that this method is more accurate and more sensitive in estimating the altitude.Besides,with this algorithm,the application sphere of the individual-soldier navigation system is expanded effectively,being the foundation of the self-adaptive individual-soldier navigation system.(5)The hardware of the individual-soldier navigation system is designed.To testify the autonomous Individual-soldier navigation system which can achieve high accuracy in a long navigation period,the comprehensive experiments involve indoor environment and outdoor environment.