一种基于模型的下肢助力外骨骼混合控制策略

A Model-based Hybrid Control Strategy for Lower Limb Exoskeleton

针对外骨骼控制中运动的灵活性和良好的辅助效果难以兼容的问题,提出了一种混合控制策略。该策略基于重力补偿、灵敏度放大和力闭环控制方法。对外骨骼的支撑腿和摆动腿采用不同的控制方法,支撑腿采用线性三维倒立摆模型进行重力补偿,结合灵敏度放大法,达到良好的跟随运动与支撑助力效果。在摆动相使用双量程传感器检测人机交互力,预测主动关节运动趋势,作为力反馈信号,构成力闭环的控制策略,满足人机随动的使用目的。完成了单关节的跟随性能测试,在此基础上进一步完成了负重蹲起试验、连续行走试验、穿戴者负重感测试,验证了外骨骼在支撑相和摆动相控制策略的有效性以及系统的助力效果和运动跟随效果。

In exoskeleton control research,it is very difficult to reconcile the flexibility of motion and good auxiliary performance.To solve this problem,a hybrid control strategy based on the gravity compensation,sensitivity amplification and force closed-loop control method was proposed.Different control methods were adopted for the support leg and swing leg of the external skeleton.The support leg adopted the linear three-dimensional inverted pendulum model for gravity compensation,and the sensitivity amplification method was combined to achieve good support effect of following motion and support.In the swing phase,a double-range sensor was used to detect the man-machine interaction force and predict the motion trend of the active joint.As a force feedback signal,the force closedloop control strategy was formed to meet the purpose of man-machine follow-up.The following performance test of the single joint was completed.On this basis,the weight-bearing squatting experiment,continuous walking experiment and the wearers'load bearing sense test were further completed,which verified the effectiveness of the exoskeleton control strategy in the support phase and the swing phase,as well as the system's support effect and movement following effect.