Detection Method For Human Dynamic Body Center Of Gravity And Gait Analysis

Clinical studies have shown that motor neurone diseasesuch as hemiplegia and stroke or the degradation of physiological functions can be reflected by gait characteristics extracted during walking.In addition,walking movements must be based on good balance ability,therefore gait information and dynamics extraction of the center-of-gravity trajectory can evaluate walking ability and rehabilitation status better,which has important academic and applied value.At present,there are still many deficiencies in this area of research,such as high equipment cost,complicated operation,and low measurement accuracy,and limited to distance.In view of the above issues,this study mainly did the following work:The inertial sensor network is constructed by using the inertial measurement unit(IMU),AVR single chip computer and 32 bit ARM microcontroller to receive the motion information from the lower limbs and torso.Based on the principle of multi-sensor information fusion,the basic structures of improved complementary filter and extended Kalman filter are established,and the attitude information of human motion joint is calculated online.The results showed that both of them have high measurement accuracy,but the PI complementary filtering algorithm is more efficient and more suitable for the system.The time-domain parameters of gait are extracted by using more simple and effective foot attitude method,and the spatial parameters such as step frequency,step size and stride length are obtained by improving the autocorrelation coefficient algorithm.The method of extracting motion information such as walking speed curve and dynamic step size is proposed.The gait detection is more reality and reliable.Taking 15 college students as the research objects,the measurement results are compared with the piezoelectric BPMS Research system.The experimental results showed that the gait parameter extraction method based on the inertial sensor network has high measurement accuracy and meets the practical needs.Using inertial sensor network,attitude information,gait information,dead-reckoning algorithm and SKC algorithm are used to detect the trajectory of human body's dynamic center of gravity.A comparative experiment was designed with 15 graduate students as objects,MOTIVE Capture system as a reference group to verify the feasibility of gait gravity measurement system based on inertial sensor network.The results show that the accuracy of this method is less than 7%.Compared with the previous measurement method,it not only greatly reduces the cost of detection,but also improves the accuracy of measurement,reduces the phase delay of the center of gravity trajectory,and is more beneficial to clinical practice application.