Research On Indoor Pedestrian Navigation Algorithm Based On Wearable Inertial Sensor

High-precision navigation and positioning systems are becoming more and more important,and navigation has become a useful tool for people to travel with.With the help of navigation tools,people can easily enjoy a large number of location-based services in the mobile media cloud.With GPS Navigation and positioning plays a key role in the outdoor,but it has a weakness is cannot penetrate indoor buildings,cause indoor positioning accuracy is poor,low accuracy,which is based on MEMS(Microelectron Mechanical Systems)technology of Inertial Pedestrian Navigation System obtained the widespread attention in recent years.Considering that the MEMS-based inertial sensors are small in size,light in weight,low-cost,convenient,and self-independent.Therefore,inertial navigation based method is applied in this research to obtain a primary navigation solutions.We know that as people age,their abilities to live decline,leading to health problems such as hearing and vision loss and memory loss.Therefore,they often need some external help in their daily activities,such as navigation system,which can effectively guide them and make their life easier and more convenient.Considering that people need to spend a lot of time in indoor places such as home,school,office building,shopping center,etc.,it is necessary to develop a wearable navigation system suitable for indoor or free living environment.This paper focuses on the research of indoor pedestrian navigation system based on wearable inertial sensor,which mainly includes the following aspects.1.This paper expounds the research background and significance,introduces the domestic and foreign research status of indoor navigation,the development of inertial pedestrian navigation system,and determines the research direction and main research content of this paper.2.This paper introduces the inertial pedestrian navigation technology,including the selection and determination of the coordinate system,the conversion relationship between the coordinate systems,and the strapdown inertial navigation algorithm.3.The zero velocity detection algorithm is proposed,the hidden markov model is introduced into the zero velocity detection algorithm,and the basic knowledge of the hidden markov model is introduced.A large number of test data are used to divide the pedestrian gait cycle into four stages.The data measured by the inertial sensor can accurately detect the stages of the pedestrian walking cycle,which solves the problem of error detection and omission detection.4.The zero-speed correction algorithm is improved,because the zero-speed detection algorithm can accurately divide the gait cycle of the pedestrian walking process,on this basis,the kalman filter and the zero-speed correction algorithm are effectively combined.According to the mutual influence of velocity error,noise error and position error,the navigation errors in the navigation system can be estimated and corrected.At the same time,an adaptive gain complementary filtering method is introduced to solve the problem that magnetometers in inertial sensors are easily interfered by local magnetic field changes,and the precision of direction estimation is improved.5.Experiments have been carried out to verify the accuracy of the improved algorithm.Tests have been carried out for three specific indoor scenes of linear path,complex figure 8 path and stair path.The experimental data has been processed and simulated by MATLAB software.