Research On Key Technology Of Pedestrian Navigation System Based On MEMS Component

Pedestrian navigation system refers to the technology of navigation and positioning for individual soldier or pedestrian.In order to improve the reliability and precision of pedestrian navigation system which works in multiple envirnments,a calibration methord based on adaptive robust least squares estimation for the accelerometer and magnetometer is proposed,and a fault-tolerant federated Kalman filtering algorithm for pedestrian navigation system composed of strapdown inertial navigation system(SINS),GPS(Global Position System)and magnetometer is studied.The main contents of the study include:Firstly,the current research status of pedestrian navigation system is fully investigated,the MEMS(Micro Electronic Mechanical System)sensor,gait analysis and zero velocity update algorithm framework are summarized.According to the current research status,the research direction of pedestrian navigation technology based on MEMS components using SINS as the core subsystem to fuse other sensors is determined.Secondly,a brief introduction to the MEMS accelerometer,MEMS gyroscope,GPS and magnetometer in the pedestrian navigation system is made.The error model of accelerometer and magnetometer is analyzed emphatically,and the calibration method of adaptive robust least squares parameter estimation is put forward,and the calibration method is tested by an experiment.Thirdly,the zero speed detector based on generalized likelihood ratio detection is described.The formula of it is derived,and the acceleration moving variance,acceleration magnitude and angular rate energy detector are introduced.A zero speed update algorithm based on the SINS error propagation model is designed,and the validity of the zero speed detector is verified by the straight line walking experiment.Fourthly,in view of pedestrian navigation system composed of multi sensors,the Kalman filtering model combined with SINS and other sensors is introduced.The fault detection methods of residual detection are emphatically expounded.The fault tolerant federated Kalman filter structure based on fualt detection method of residual detection is designed.At last,in order to test the feasibility and accuracy of pedestrian navigation algorithm,the fault tolerant federated Kalman filtering algorithm is simulated.From the simulation results,it can be found that the algorithm can isolate the faults,make the system avoid the influence of faults,and ensure the continuity and correctness of the filtering.Indoor and outdoor experiments were carried out with MEMS components installed on foot.The experimental results show that the positioning error of pedestrian navigation system based on MEMS component is less than 3%.