Research On Lower Extremity Exoskeleton Control Technology Based On Human-machine Interaction

Lower extremity exoskeleton is a wearable intelligent mechatronics system,which owns a wide range of application scenarios and is a significant filed in the research of robotics.Exoskeletons have been applied into military,aerospace,industrial,civil,medical and other fields.The application in the medical field has achieved good results and has been put into industrialization.However,due to its complex system structure and human-machine interaction characteristics,many problems need to be solved.Therefore,this paper mainly studies the problem of the control algorithm for assist exoskeleton.This paper completes structural optimization,sensor system design,electrical system construction,and builds a software system with efficient data structure and process tasks to ensure the reliability of the task operation and scalability of the function.And then we analyze the physiological characteristics of human motion,and use the combination of two basic motion modes,support phase and swing phase,to describe the complex motion behavior of human body.Based on the GRF gait identification method,we invent a flexible pressure shoe that can be adapted to complex road surfaces and has the function of both gait recognition and motion intent recognition.Finally,we use the fuzzy logic to process the plantar pressure data to get the motion gait.In addition,using the PID form of admittance theory,an end force-velocity model is established to describe the relationship between human-machine interaction and human motion intent.Through the kinematics and dynamics analysis of different motion phases,an experiment to identify the parameters of the model at any phase is designed.Combined with all the above research results,establish the overall control strategy of exoskeleton.The PID controller is used in the swing phase,the MFAC controller is used in the support phase,and the gravity coefficient is proposed to adjust the gravity in the phase switching process to achieve smooth switching.Finally,the experiment of exoskeleton assist effect is designed,and four experiments such as swinging,lifting,walking and going up the stairs are finished.The effectiveness of this method is verified by analyzing and researching the experimental results.