Research On Multi-source Information Enhanced Pedestrian Positioning Technology Based On MEMS Inertia Sensors

With the increase of people's indoor activities,it is critical to accurately obtain indoor pedestrian positioning information.The personnel positioning technology based on micro-electro-mechanical inertial sensor has been widely used because of its strong autonomy and good concealment.However,the positioning error of this technology will accumulate over time,which is difficult to meet the long-term high-precision positioning requirements.Therefore,some scholars have proposed multi-source information enhancement personnel positioning technology,which combines multiple information sources using optimization criteria to reduce the positioning error of a single information source.In order to solve the problem of inertial positioning error accumulating over time,this paper regards inertial positioning as the main information source and combines multisource information to enhance the plane position,altitude and heading respectively,which effectively improves the positioning accuracy.The main research contents are as follows:Firstly,for the continuous stable scene of multi-source information,the state equation and the observation equation are established for the positioning information of inertial positioning and multi-source information,and the position enhancement is performed by Extend Kalman Filter.For the multi-source information missing scene,a sparse multi-source information location enhancement algorithm based on fault-tolerant decision tree is proposed.The position information of inertial positioning and multisource single-point positioning data information are used as a common reference to construct a fault-tolerant decision tree model for sparse information.The source data is used to determine the reliability of the fault-tolerant decision tree and perform weighted fusion to achieve position enhancement.The simulation results show that the proposed algorithm can enhance the position of inertial positioning quickly and effectively,reduce the position error and improve the positioning accuracy.Secondly,aiming at the problem of height error,a height enhancement algorithm based on feature constraints is proposed,which combines accelerometer and barometer to identify the running state of upstairs and downstairs,and uses this information to correct the height to improve the accuracy of the system.Experiments show that the relative accuracy of the algorithm is 96.95%,and it can locate the floor effectively.Finally,aiming at the problem of accumulated errors in gyroscope solution and course fluctuation in pedestrian direct travel,this paper carries out direct travel judgment based on different movement states of personnel.The improved heuristic drift elimination algorithm and weighted fusion algorithm are used to modify the current heading by feedback,and then the heading enhancement is realized.The experimental results show that the multi-source information-enhanced inertial positioning system studied in this paper can achieve continuous and reliable positioning under multi-source positioning information and provide accurate and lasting positioning information.The positioning error is less than 1.5%,and the effective positioning time is more than 30 minutes.