Nowadays,with increasing attention to issues such as population aging and laborintensive industrial transformation,various exoskeletons that help the elderly and the disabled and enhance lower limb functions have been widely used and researched in medical rehabilitation,industrial production and military fields.However,the design and research of lower extremity exoskeletons still need to solve problems such as low accuracy,poor control performance,uncoordinated human-machine coordination,and unintelligent control strategies.Therefore,the goal of this thesis is to study the coupling system composed of the lower extremity exoskeleton and the human body.Starting from the accuracy,safety and intelligence of human-machine coordinated control,relevant research is carried out in order to achieve efficient control goals.The main research work of this thesis includes:(1)The dynamic model of the two-degree-of-freedom lower extremity exoskeleton was constructed,and linearized according to its structural characteristics,and a more accurate dynamic model was obtained.A backstepping controller was applied in the inner loop of the first-generation exoskeleton controller,and a backstepping controller with finite time-based convergence was further used,and the controller was used in the secondgeneration exoskeleton;(2)Based on the first-generation robotic lower extremity exoskeleton system,a sliding-mode disturbance observer with finite-time convergence is designed.The control goal is to design a sliding-mode disturbance observer control based on finite-time convergence,so that the output of the system is saturated when the input is saturated and the control is singular.In the presence of external disturbances and external disturbances,it follows the given expected output of the nonlinear system,and achieves a better tracking effect.(3)In order to adapt to the human-machine coupling strength of the exoskeleton at different synchronous frequencies and reduce the human-machine confrontation during the movement,the variable admittance control is realized on the basis of the upper fixed admittance control.The virtual damping and virtual stiffness are traversed at different synchronous frequencies,and the human-machine coupling torque information,position information,etc.are collected.And after multi-dimensional analysis of the experimental data,the man-machine coupling index is proposed.At each step frequency,the virtual admittance value corresponding to the best man-machine coupling index is optimized,and then the virtual admittance parameters of each step frequency are fitted,and the fitting curve is obtained and applied to the variable admittance controller.To verify the effectiveness of the step frequency-based variable admittance control strategy,the second-generation experimental prototype was used to carry out relevant verification experiments. |