摘要
为根据患者需求调整训练模式,提升康复机器人的个体适应性和应用效果,以下肢康复机器人为例,结合人机交互运动的协调一致性,将人机视为一个整体,简化为连杆和运动副的组合体,依据膝关节、髋关节及裸关节在矢妆面内的变化,引入拉格朗日方程对人机系统进行动力学建模;同时,融合sEMG和足底压力的特征,构建自适应性的人机交互控制策略,以实现主动、被动式训练的灵活调整;最后,采用虚拟样机技术,融合Pro/E、Mechanism/Pro及Adams软件联合构建康复机器人的虚拟样机,透过给定期望轨迹与实际轨迹的差异比较,分析康复机器人建模方法的可行性,并基于sEMG、和人机交互力等多源信号的运动识别算法,通过运动仿真实验,进行人体运动意图的实时解码。仿真结果表明,ADAMS仿真与运动学模型所得的驱动力矩基本一致,验证了该康复机器人设计的可行性,且基于sEMG信号的采集与分析,可准确识别人体运动意图,正确率在90%以上,可为针对性的康复训练提供有效支撑。
In order to adjust the training mode according to the needs of patients and improve the individual adaptability and application effect of the rehabilitation robot,this paper took the rehabilitation robot of the lower limbs as an example,combined with the coordination and consistency of human-computer interaction movement,regards the human-computer as a whole,simplified as the combination of the connecting rod and the motion pair.According to the changes of the knee joint,the hip joint and the bare joint in the sagittal plane,the Lagrange equation is introduced to model the human-computer system dynamics.Based on the characteristics of sEMG and plantar pressure,a self-adaptive human-computer interaction control strategy was constructed to realize the flexible adjustment of active and passive training.Finally,the virtual prototype of rehabilitation robot was constructed by combining Pro/E,mechanism/pro and ADAMS software,the feasibility of the modeling method of rehabilitation robot was analyzed by comparing the difference between the expected trajectory and the actual trajectory.Based on the motion recognition algorithm of multi-source signals such as sEMG and human-computer interaction,the human motion was carried out through the motion simulation experiment Real time decoding of intention.The simulation results show that the driving torque obtained by ADAMS simulation and kinematic model is basically the same,which verifies the feasibility of the design of the rehabilitation robot.Based on the collection and analysis of sEMG signal,it can accurately identify the motion intention of human body,and the accuracy is more than 90%,which can provide effective support for targeted rehabilitation training.
作者
黄智晖
Huang Zhihui(School of Education,Xi′an Universiy of Arts and Science,Xi′an 710065,China)
出处
《电子测量技术》
2020年第9期127-132,共6页
Electronic Measurement Technology
