The Design Of Gait Detection System Based On Multiple AHRS Nodes

Gait detection is an important subject in the field of humanoid robot.With the development of science and technology,gait detection device and analysis system is more and more,such as plantar pressure detection device,surface EMG detection device,force platform and vision-based gait analysis system.In this paper,a dynamic capture system based on AHRS(Attitude and Heading Reference System)is designed to realize real-time detection of gait.The main research contents are as follows:(1)Analyzing system error based on Allan Variance method.The characteristics of the different inertial sensors vary widely.In this paper,Allan Variance analysis method is used to analyze the error of several commonly used nine-axis inertial sensors(such as MPU9250,LSM9DS0 and BMX055).The MPU9250 with the smaller error was selected as the system chip by analyzing the measurement data.(2)Designing and networking of multi AHRS nodes.The hardware design of a single AHRS node based on STM32F405 and MPU9250.By comparing several typical wireless transmission technology,the data transmission mode based on Wi-Fi network is designed,which can meet the requirements of data transmission and improve the system's flexibility.(3)The design of communication and data processing program.Reading the data from AHRS node by I~2C Interface.The Socket communication in Wi-Fi network is designed.This paper presents a kind of attitude estimation method based on gradient descent algorithm,and completes the hardware and software platform of human gait detection system.(4)Human body modeling and linkage based on Virtual Robot Experiment Platform(V-REP).The human body model is constructed in V-REP,each node is corresponding to the human body,and the motion of the hip joint and the knee joint is eliminated by using the kinematics of the robot.Human motion data are collected in real time by the AHRS which is installed in the main part of human lower limb,and the collected gait data is transporting to V-REP to reconstruct human body movement posture,which achieve the real-time capture of the human body gait.In this paper,the human gait detection system is constructed based on multiple AHRS nodes.Preliminary experiments have been designed and completed in the laboratory through the analysis of sensors,hardware design and networking,data transmission and settlement,as well as the reconstruction of the human models,etc.