Research On Upper Extremity Exoskeleton Rehabilitation Training System Based On Surface Electromyographic Signal Control

With the continuous growth of our national economy,the aging problem of population is more and more serious,and the elderly patients with hemiplegia and disability are increasing.For patients with hemiplegia,the traditional rehabilitation training,including physical rehabilitation training based on rehabilitation technicians,has problems such as long cycle,large personnel consumption and limited effect,which will inevitably affect the effect of rehabilitation training.Therefore,the research and application of upper limb rehabilitation robot technology has gradually become the focus of many rehabilitation scholars at home and abroad.In this paper,an upper limb exoskeleton rehabilitation robot integrated with variable stiffness device is designed.The driving stiffness can be adjusted in time,and the robot can be controlled by EMG signal,which greatly improves the safety of human-computer interaction.Firstly,by analyzing the structure and motion characteristics of the human upper limb,the overall structure of the robot was designed by using Solidworks software,the driving mode and motor selection of the upper limb rehabilitation robot were determined,and the principle of the variable stiffness device was analyzed.Next,the surface EMG signals of the relevant muscles of the upper limbs were collected and denoised by Butterworth bandpass filter.After denoising,the time domain features of the surface EMG signals were extracted by sliding window method.The action classification prediction model based on principal component analysis,BP neural network algorithm and support vector machine algorithm was established,and the extracted eigenvalues were used as inputs to classify the actions.Finally,the variable stiffness experiment platform and the upper limb rehabilitation robot system experiment platform were built,including EMG acquisition module,control module,drive module,upper computer processing module and exoskeleton robot module.The stiffness measurement experiment was completed,which verified that the designed variable stiffness device could freely adjust the stiffness to improve the intensity of rehabilitation training.The motion Angle tracking experiment of the exoskeleton robot was completed,which verified the effectiveness and feasibility of the designed exoskeleton robot,and the control accuracy experiment of EMG signal was completed.It is verified that the control method of the exoskeleton robot using the EMG signal of the upper limb has good accuracy and real-time performance,and can carry out rehabilitation training according to the movement intention of the patient.The upper limb exoskeleton robot system designed in this paper based on s EMG control focuses on the design of a variable stiffness device and the improvement of the robot’s human-computer interaction ability,which increases the patients’ rehabilitation initiative and enthusiasm.