Research On Indoor Positioning Algorithms Based On AHRS

Micro-Electro-Mechanical System(MEMS)technology plays a key role in mobile devices.Due to its small size and low power consumption,it is widely used in navigation and positioning,robots,UAVs and other fields.Independent system composed of MEMS cannot achieve accurate positioning in the environment of satellite signal rejection,and its positioning errors will accumulate over time.The positioning system based on Attitude and Heading Reference System(AHRS)can use gait as a priori information to reduce errors and achieve short-term indoor high-precision positioning,which is a research hotspot at present.In this paper,the positioning of a single AHRS adaptive different wearing position is taken as the core,and the topology of the human body is recognized and developed.The positioning method of the switchable AHRS wearing position is realized.The specific research contents are as follows:1.This paper proposes the Body Topology Recognition(BTR)algorithm.Generally,in a positioning system that uses MEMS device-based AHRS for navigation,the AHRS wears different positions,and the key parameters in the same algorithm are different.Typical wearing positions include the instep,ankle,calf,waist,head,and wrist.These positions not only reflect the movement characteristics of the person's walking process,but also are easy to wear.By acquiring the data of six typical position sensors of pedestrians in walking state,a feature construction method based on gait information is designed,and a classifier based on Gradient Boosting Decision Tree(GBDT)is constructed to introduce real-time detection.Completed the design of the entire BTR algorithm.The method can recognize the six typical wearing positions accurately,and the real-time performance can be controlled within 1.2 seconds.2.This paper proposes a gait detection algorithm for different wearing positions.Through the association between the motion characteristics and gait of different wearing positions,this paper divides the human motion area into three large categories,including the lower limbs,trunk and hands.According to the motion characteristics of the three types of positions,three gait detection algorithms are proposed,which can be used in conjunction with the recognition results of BTR algorithm with high flexibility.3.This paper proposes a step-length compensation method.During the pedestrian walking process,when the AHRS wearing position is switched,the BTR algorithm is used to recognize the human body topology,which will occupy the 1.2s duration of the switched state,which leads to the missing step in the trajectory drawing.For this problem,the paper draws on the trajectory.The step-length compensation is introduced to ensure the continuity and completeness of the trajectory drawing.The effectiveness and repeatability of the proposed algorithm are verified by experiments.The experimental results show that the proposed BTR algorithm can accurately recognize six typical wearing positions with an accuracy of 99.11% and a gait detection accuracy of 99.40%.Most importantly,this paper combines BTR algorithm,gait detection algorithm and step-length compensation method to realize the accurate positioning of pedestrian walking process,which ensures the continuity and real-time of trajectory rendering,and shortdistance indoor positioning accuracy can achieve less than 2% distance error and end-to-end error.This breaks through the limitation of single position wear of traditional positioning module,realizes the flexibility,diversity and friendliness of its wear.It can be widely used in human or robot attitude calculation,posture recognition,trajectory estimation and other directions,and has high application value and application prospects.