Research On IMU-Based Inverse Dynamics Modeling Of Human Walking And Assistance Simulation And Experiment

At present,there is a growing concern about that using soft exoskeletons to assist human body in daily life to reduce metabolize rate.The basis of soft exoskeletons to assist human is the joint moment curves of the human body when walking normally.However,in most current studies,it is difficult to solve joint moments of the experimenters.Therefore,researchers can only make soft exoskeletons work according to the moment curves from statistical law,which limits the performance of soft exoskeletons.Fist of all,a human body kinematics model based on inertial measurement unit(IMU)is established,which is used to solve the kinematics information of the human body while walking.Based on the analysis of human motion characteristics and IMU signal characteristics,the kinematics construct equation between IMU and human lower limb joints is established.And the construct equation is solved based on optimization algorithm,so that IMUs can be aligned with the limb segments automatically,avoiding the cumbersome operation of traditional methods.On this basis,the human kinematics information is sovled based on angular velocity and acceleration respectively,and the data is optimized based on the data fusion algorithm.The accuracy and stability of the model were verified using optical motion capture system.Secondly,after solving the motion information,a human body walking inverse dynamics model based on IMU is established.The model simplifies the movement of the human body into the sagittal plane,and determines the inertia and geometric parameters of each limb segment of the human body based on statistical laws.At the same time,combined with "smooth transition" characteristics of the ground reaction forces,the model makes the upper limb centroid as benchmark to establish the motion and force equations of various limb segments of the human body.On this basis,using the NewtonEulerian method,the human body's internal forces and joint moments are solved respectively.The accuracy and stability of the model were verified in conjunction with an optical motion capture system and a force tableThen,for the effect of using soft exoskeletons assist human based on the inverse dynamic model proposed by this paper.The general human body model in the simulation software is scaled to obtain the human body model of experimenter.Combined with the optimization algorithm,the scaled human body model and the measured kinematics data are optimized to reduce the simulation error.On this basis,for the metabolic rate when human walking normally,assisted based on method and assisted based on method of this paper,a simulation was performed.The results show that when the human body is assisted by the method of this paper,the human body has the least metabolic rate.Finally,in order to verify the effect of the method of this paper,the e metabolic rate of human body in the above three states was examined.The experiment results also show when the human body is assisted by the method of this paper,the human body has the least metabolic rate.