| The advancement of science and technology and the development of society have promoted the increasing demand for the reliability of high-precision positioning and navigation.Global Positioning System(GPS)fully meets people's needs in the outdoor environment,but the indoor environment is complex and changeable.A large number of walls and obstacles make the GPS signal decay rapidly,resulting in the indoor GPS signal being extremely weak and unable to perform normal positioning.Micro-Electro-Mechanical System(MEMS)sensor chips have the advantages of small size,light weight,low power consumption and low cost.Wearable Inertial/Magnetic Measurement Unit(IMMU)based on MEMS sensor chip can complete indoor positioning independently and has high positioning accuracy in the short term.It has great application value in the fields of medical supervision,public security,intelligent space and military defense.This paper implements Pedestrian Dead Reckoning(PDR)based on wearable IMMU.Due to the limitation of the manufacturing process,the MEMS sensor itself has drift errors,which will greatly affect the positioning accuracy.In order to effectively reduce the influence of drift errors and improve positioning accuracy,this paper conducts the following research:1.This paper develops a wearable pedestrian positioning system based on STM32F405 and MPU9250,and completes the schematic diagram and Printed Circuit Board(PCB)design.The wearable pedestrian positioning system is fixed to the pedestrian instep for experimental data collection,and the experimental data is filtered and calibrated.2.In this paper,PDR is implemented in four steps: step detection,gait phase detection,step length estimation and attitude solution.Firstly,the motion model is constructed by making full use of the leg opening angle information during pedestrian movement,and a novel method is proposed to perform step detection based on the pitch obtained.Then,the gait is detected by the acceleration and angular velocity,and the stance phase and swing phase in each complete gait are distinguished.The gait phase detection algorithm is the basis for step length estimation and attitude solution.Next,the step length is estimated according to the improved(Zero Velocity Update)ZUPT algorithm.When the pedestrian is in stance phase,the velocity is zero,and the drift error of the acceleration value in each step is assumed to beconstant(this value is generally different at different steps),so as to calibrate the velocity,and then integrate to obtain the estimated step size.Finally,it is assumed that the drift error of the angular velocity value at each step cycle is a fixed value(the value of different step cycles is generally different).An improved complementary filtering algorithm is proposed to calibrate the attitude based on the complementary characteristics of accurate response of the accelerometer under low dynamic condition and good performance of the gyroscope under high dynamic condition,and combined with the output data of magnetometer.3.In order to verify the feasibility of the presented PDR algorithm,the paper repeats the experiments in indoor environments to obtain experimental data,and then uses the Matlab mathematical platform for simulation verification.The experimental results show that the average error of the step detection algorithm proposed in this paper is 0.42% under the condition of pedestrian walking and running.In addition,the average accuracy of gait phase detection based on acceleration and angular velocity is 99.46%.The average distance error of the PDR algorithm proposed in this paper is 1.36% in indoor environments,and the average end-to-end error is 1.24%,which are all within 1.5%,indicating that the proposed PDR algorithm has a good positioning effect. |
PDF Full Text Request