Research On Pedestrian Navigation And Positioning System Based On Multi-sensor Zero Velocity Correction

With the rapid development of social economy,the range of people's activities is gradually expanding.Not only in the square,street,yard and other outdoor areas,but also they often appear in the indoor environment of bridges and tunnels,large hospitals,shopping malls,underground parking lots and so on.However,it is impossible to locate its own position,due to no wireless navigation signal,such as GPS and BDS in the indoor environment.In order to solve this problem,the position tracking and navigation services are carried out by using the MEMS inertial sensor which does not depend on the external signal as the autonomous navigation component.Indeed,due to the zero drift error,motion state error and navigation solution error of the MEMS inertial sensor,the position and navigation information are greatly deviated.Aiming at the error caused by inertial sensors navigation during pedestrian walking,we proposed a zero velocity correction algorithm based on multi-sensor data fusion.Firstly,the inertial measurement module is bundled on the pedestrian foot surface to acquire the acceleration and angular velocity parameters of pedestrian travel in real time.Then,by means of the SVM zero velocity detection classification method,the moving and static state can be identified quickly and accurately.In the stationary phase velocity and angular velocity is zero.At the same time,the magnetometer and the barometer are used to determine the azimuth and altitude respectively,so as to correct the parameters,such as the velocity,angular velocity,azimuth and altitude of the pedestrian.Finally,the pedestrian walking error is filtered by the extended Kalman filter method,the position and direction information of the inertial navigation solution is compensated in real time to track the pedestrian trajectory.In order to verify the effectiveness of the design solution,actual walking navigation and positioning experiments were conducted.According to analyze of the experimental data,the results of multiple measurements show that the pedestrian trajectory can be completely consistent with the specified path,and the maximum position error is less than 3%.Therefore,this study can achieve accurate tracking and positioning of pedestrian walking trajectory,which is of great significance for navigation and location services in the absence of wireless signals.