Exoskeleton EMG-admittance Control Technology And Application Based On Biomechanical Model

With the continuous development of aging in our country,there are more and more patients with lower extremity movement disorders.Lower limb rehabilitation exoskeleton robots can effectively help patients with lower limb movement disorders to restore their walking function and improve their quality of life.Among the control methods of lower limb exoskeletons,the existing methods mainly include physical signals such as force feedback and biological signal methods such as surface electromyography signals.With regard to the former method,due to the hysteresis of the signal,it is difficult to identify and feedback the wearer’s active movement intention in time.As for the latter method,the control of the exoskeleton is not precise enough at present.Generally speaking,the method of pattern recognition and machine learning is used to divide the movement into different movement modes,then collect the corresponding sEMG signals and train the model.Corresponding control strategies are adopted for different motion modes,which are not precise enough and have poor adaptability.In this paper,based on the sEMG signal,the joint torque estimation will be realized through the biomechanical model,and the adaptive admittance will be used to realize the control of the exoskeleton.The main contents are as follows:（1）A multi-mode sensing knee exoskeleton platform is built,which integrates motion systems,sensing systems,control systems,and communication systems.It can collect human surface EMG signals,angle data,and exoskeleton motor data in real time.Real-time control of the exoskeleton.Its core is a control system based on Labview.（2）In order to realize the good interaction between the lower limb exoskeleton and the wearer,an admittance control model based on the interaction torque is designed.The influence of the admittance model parameters on the control effect is analyzed,and the adaptive admittance control rate is designed,which can adjust the admittance parameters in time with the change of the wearer’s swinging leg speed and the change of the wearer’s interaction torque with the exoskeleton.And according to the periodicity of motion,a neural network is proposed to further optimize the admittance control rate.Finally,simulation experiments are carried out to verify the excellent effect of adaptive admittance control in exoskeleton control.（3）Aiming at the complex and poor effect of estimating joint torque based on surface electromyography,an efficient method based on biomechanical model was proposed to estimate the active torque of knee joint by using surface electromyography.With the aid of exoskeleton,the dynamic model of the calf swinging around the knee joint can be obtained simply and relatively accurately.In addition,a method for parameter identification of Hill muscle model is proposed and implemented,which realizes the effective estimation of the active torque of the knee joint by the surface EMG signal.Finally,through experiments,the effectiveness and feasibility of the proposed knee joint torque estimation method are verified.（4）For the real-time refined exoskeleton control based on surface electromyography,a real-time control of exoskeleton position based on surface electromyographytorque mapping is proposed.The joint torque estimated by surface electromyography is combined with adaptive admittance control,and a leg A fine active teleoperation control method of the exoskeleton based on the surface electromyography signal.The method can not only drive the exoskeleton to move when the person is actively moving,and realize the assistance,but also realize the effective support of the exoskeleton when the person is passively moving.The EMG-admittance exoskeleton control method based on the biomechanical model proposed in this paper realizes the finer control of the exoskeleton through EMG,and uses the EMG characteristics to realize the passive support of the exoskeleton,which expands the application scenarios and provides a better solution for the rehabilitation of the lower limbs.Exoskeleton control research provides a new control idea.