Research On The Structure Design And EEG Signal Processing Of Lower Limb Rehabilitation Robo

There are more and more patients with impaired lower limb function in China,but the number of rehabilitation physicians is far from meeting the demand,so the design and research of lower limb rehabilitation robots is of great significance to improve the rehabilitation quality of patients and reduce the burden on patients’ families.Aiming at the problems of complex structure and single training mode in lower limb rehabilitation robots,this paper proposes a two-degree-of-freedom lower limb rehabilitation robot.The modular structure design enables the robot to realize two rehabilitation training exercise modes: leg swing and ground walking;The research of EEG signal processing has promoted the practical application of EEG signals in the control of lower limb rehabilitation robots.Based on the rehabilitation goals and medical principles that meet the patient’s daily activities,the physiological structure of the human lower limb was analyzed,and the design parameters of the robot were determined according to the size standard in ergonomics.The overall structure of the lower limb rehabilitation robot was designed,and the leg length adjustment mechanism and hip width adjustment device could adapt to patients with different postures through manual operation;The lifting platform can drive the mechanical leg to move up and down,and realize the robot to switch between the air swing leg movement mode and the ground walking movement mode;The combination of mechanical limit devices,joint module limit protection and emergency stop switches ensures patient safety.Solid Works was used to build a virtual prototype of a lower limb rehabilitation robot and perform strength checks on key parts.ADAMS was used to simulate and analyze the two motion modes of aerial leg swing and ground walking of the rehabilitation robot,and the structural design of the robot met the action requirements of rehabilitation training.According to the analysis comparison,the Lagrange function method is applied to solve the dynamic equation of the humanmachine pendulum model.Considering that the friction force affects the control accuracy when the mechanical leg joint rotates,this paper proposes a trajectory tracking control of the mechanical leg based on fuzzy compensation.The fuzzy algorithm is used to fuzzy compensate the nonlinear friction,and the control rate of the mechanical leg system is designed by inverting the control method.Simulation results show that the introduction of fuzzy compensation can effectively overcome the influence of friction and improve the control accuracy of the system.The 20-pass EEG acquisition device of NE Company in Spain was used for the experiment.According to the paradigm of EEG acquisition experiment,the specific operation process of the experiment was designed,and the EEG signals of left leg movement and right leg movement were collected.For the original EEG data,Butterworth bandpass filtering was used to remove the noise,and independent component analysis was used to eliminate the ocular artifact.Secondly,wavelet packet decomposition is used to divide the frequency band of the filtered data,and the common space pattern algorithm is used to extract the signal features.Finally,three kinds of classifiers(support vector machine,random forest and convolutional neural network)are used to classify and recognize the EEG signals respectively,and the best convolutional neural network is selected as the classifier in this experiment.In this paper,a lower limb rehabilitation robot with two motion modes is designed,a three-dimensional model of the overall structure of the robot is established,a trajectory tracking control algorithm of fuzzy compensation for mechanical legs is proposed,and EEG data of two kinds of motion imagination are collected and processed,and a suitable classifier is selected,which lays a foundation for the development and control of robot prototype.