Passive&Active Back Support Exoskeleton Design And Control Strategy Research

In the fields of logistics,construction,and industry,workers are engaged in physical labor with the high-intensity loading which increases the risk of muscle and bone diseases.With the intensification of the aging issue in China,how to prolong the occupational life of workers and reduce the incidence of occupational diseases is becoming urgent social topic.Studies shows that exoskeletons can improve the wearer’s athletic ability and reduce the stress on human muscles and bones,thereby reduce the incidence of muscle and bone diseases.At present,most kind of back support exoskeletons are lumbar assistance,few studies focus on lumbar assistance together with knee compression during lifting.To fill this gap,this research designs an active hip joint and passive knee joint lifting exoskeleton.It not only provides assistance to the wearer’s waist,but reduces knee joint compression by transmitting the weight of human trunk and objects to the ground through the exoskeleton thigh rod and knee joint.The main contents of the paper are as follows:Combined the design goals of light weight,comfort,and practicability with scenarios,the overall scheme was determined including active and passive combination,one-way torque transmission,and gravity support.Based on human anatomy and bio-mechanical characteristics,the layout and degrees of freedom are established around the ergonomic configuration.Based on the layout and modular ideas,the hip joint,back support,knee joint,and pressure shoes are designed in detail.Finally,to grantee the key parts strength,components simulation analysis is done.Exoskeleton sensing control system is designed with STM32 as the controller.By analyzing the intention of the human body,intention recognition module is established which includes IMU(Inertial Measurement Unit),foot pressure sensors,and hip torque sensors.Based on the real-time control requirements,servo drive module is designed which is mainly based on integrated brushless motor of the hip and DC gear motor of the knee.Based on the hardware resource configuration of the sensor control system,the design of the embedded software system of the exoskeleton which includes ADC,IIC and CAN ports is carried out.Intent recognition strategy with FSM(Finite State Machine)is presented as upper level control strategy and the low control algorithm is based on an admittance model for exoskeleton following and assistance.To identify the walking and squatting&stooping phase,the article analyzes the laws of walking,squatting,stooping and fuse feet pressure signal.Based on above Information,we design walking phases judgment rule with fuzzy algorithm and lifting phases judgment rules with back IMU posture.FSM phase switching rules for gait recognition is achieved by combination fuzzy algorithm and the IMU posture.To achieve precise control of exoskeleton,exoskeleton support and swing model is deduced with Lagrangian Equation.Admittance model of the man back with exoskeleton is established for better following.The following and power-assisted algorithms were designed with admittance model as the outer loop and PD position controller as the inner loop.The admittance parameter adjustment rule which is for movement error reducing is established.The variable admittance control effect is verified by Matlab simulation.Finally the article set experiments to evaluate exoskeleton system and control algorithm.Set the mechanism motion experiment,single joint follow and exoskeleton follow experiment,walking support,and handling assistance experiment to confirm the system recognition accuracy&real-time performance.Design the gravity support and breathing efficiency experiments with and without exoskeleton,evaluate the exoskeleton support and power-assisting efficiency.The experimental results show that wearing the exoskeleton can achieve good support and power-assisting effects which achieves waist assistance and reduce knee compression design goals.