Research On The Design And Control System Of Knee Assisted Exoskeleton

Exoskeleton has a wide range of application prospects in the fields of national defense and people’s livelihood,helping the elderly and disabled.Rigid exoskeleton can output large power,but its mass is large and its flexibility is low.In the process of movement,the rotation center of the exoskeleton joint and the human joint may not align,and the rigid structure will produce large unintended torque on the human skin.The soft exosuit is comfortable to wear,light in weight,and flexible.However,it does not have a fixed rigid structure to support and transfer auxiliary force,so it cannot complete the task of assisting in some scenarios.At the same time,domestic and foreign scholars mainly study the lower extremity exoskeleton power-assisted joints for the hip and ankle,and there are few studies on the knee power.However,with the increase in the proportion of the elderly population in China,more and more elderly people are faced with knee injury,lack of muscle strength around the joint and other problems.To solve these problems,a new soft knee exosuit is proposed in this paper,which is used to assist the elderly in sit-stand conversion,flat and uphill walking and other activities.Vicon motion capture system and Visual 3D simulation were used to obtain the kinematic and dynamic parameters of human body walking at different slopes,which provided data support for the design of expected power curve of exoskeleton working at different slopes.According to the motion characteristics of the knee,two knee exoskeleton systems were designed and assembled,the output torque of the rigid knee exoskeleton was analyzed and calculated,and the fixed anchor point and force transfer path of the soft exosuit for knee were designed.The performance of the two exoskeletons were evaluated,and the soft exosuit for knee was selected for control strategy study.In order to achieve precise assisting,the expected assisting curves of the knee were designed according to the kinematic and dynamic parameters of the knee,and the Bowden line contraction model was constructed to obtain the relationship between the knee Angle and the Bowden line contraction.A control system combining the admittance controller and the iterative learning controller was constructed.The admittance controller was used to realize the real-time tracking of the actual power curve of the exoskeleton to the expected power curve,and the iterative learning controller was used to generate the compensation line length iteratively,eliminating the errors caused by the difference of the wearer’s physical signs and binding wearing.In order to evaluate the assisting effect of soft exosuit for knee,auxiliary force tracking experiment,metabolism experiment and muscle fatigue experiment were conducted.In the auxiliary force tracking experiment,after 20 rounds of control optimization,the auxiliary force curve of the soft exosuit for knee showed good tracking performance to the expected power curve.The results of metabolism experiment of 6 subjects showed that the average net metabolic consumption was reduced by 4.61%,10.76% and 12.49%,respectively,when wearing the soft exosuit for knee at 0%,15% and 25% slope.The results of muscle fatigue experiment showed that when 6 subjects wore the soft exosuit for knee and walked at different inclines,the fatigue of knee activity-related muscles decreased to different degrees.