不同减小运动学和动力学数据误差的方法对下蹲跳运动仿真结果的影响

探究不同方法处理运动学和动力学数据对运动仿真计算结果的影响,进而减小数据误差以提高运动仿真计算结果的准确性。采集1名普通男性在校大学生的一次下蹲跳(CMJ)测试的运动学和地面反作用力(GRF)数据,分别使用直接计算方法(DK)和逆向运动学方法(IK)计算关节运动学数据;对原始GRF数据(RF)分别进行10 Hz(SF)和50 Hz(HF)的Butterworth 4阶低通滤波处理。应用Open Sim 3. 3软件结合不同方式处理的运动学和GRF数据进行CMJ运动仿真,计算CMJ过程中下肢肌肉-肌腱长度变化和下肢各关节力矩。采用复相关系数(CMC)和平均绝对变化值(MAV)来评价结果的相似性。结果显示:(1)DK和IK计算的CMJ过程的关节角度曲线相似度高,CMC值为0. 93～1. 00,但两种方法计算的髋关节内收外展角及旋转角结果间差异较大,MAV值分别为4. 40°和4. 81°;(2)DK计算的下肢环节长度最大变化量达48 mm,IK计算的下肢环节长度在CMJ过程中保持不变;(3)两种方法计算的肌肉-肌腱长度相似度与各关节角度结果相关;(4)RF、SF和HF计算所得的下肢关节力矩结果相似,但落地缓冲阶段结果差异较大。所得结论如下:(1)使用IK方法处理运动学数据更有利于减小运动学数据误差,提高运动仿真结果的准确性;(2)不同截断频率低通滤波GRF数据对关节力矩计算结果的准确性有影响。建议根据运动特征,选择不同的截断频率处理GRF数据。

Effects of Different Methods to Decrease Kinematic and Kinetic Errors on Simulation Results of Counter Movement Jump

Objective: The purpose of this study was to quantify the effects of different methods processing kinematics and kinetics on simulation results, in order to decrease errors and improve accuracy of simulation results. Methods: Collecting the kinematics and ground reaction forces (GRF) of one male college student during counter movement jump (CMJ); calculating joint kinematics using direct kinematics (DK) and inverse kinematics (IK) respectively; and filtering raw GRF (RF) with 4th Butterworth low pass filtering of 10Hz (SF) and 50Hz (HF) cut off frequencies. Dynamic simulations of different CMJ was used to compute muscle-tendon length and joint moment in OpenSim3.3 software. The similarity of results were evaluated by coefficient of multiple correlation (CMC) and mean absolute variation (MAV). Results: (1) Although the results of joint angle are similar between DK and IK, CMC is between 0.93 and 1.00, the results of hip adduction and rotation are different, MAV is 4.40°and 4.81°respectively; (2) The peak variation of segment length is 48mm in DK, however, all segment length of IK are invariant during CMJ; (3) The results of muscle-tendon length are associated with joint angle; (4) Joint moment is similar among RF, SF and HF (CMC=1.00), which is different during landing. Conclusion: (1) Using IK is beneficial to decrease kinematic errors and to improve the accuracy of simulations. (2) Low pass filtering GRF influences the accuracy of joint moment , which suggests that the cut off frequency should be chosen according to motion characters.