Research On The Exoskeleton For Walking Assistance To The Physically Weak Persons

With the increase of the aging population in China,more and more elderly adults become physically weak persons who need extra care due to the deterioration of their body functions.And they suffered from walking and stair-climbing problems in their daily living activities.Tripping caused by walking and going upstairs and the secondary injuries caused by the tripping are aggravating the economic burden of the family and the medical burden of the country.As a kind of walking assistance device,the exoskeleton robot has unique advantages in assisting physically weak persons to walk and climb the stairs with its wearability and mobility.Therefore,the research of an exoskeleton robot that can help physically weak persons to walk and go up and down the stairs has a very positive significance to solve the current dilemma,reduce the pressure of medical care,improve the technology and information levels of the elderly service,and promote the implementation of the intelligent aging project.In this thesis,based on the anatomy of human joints and the walking movement rules of physically weak persons,an anthropomorphic walking assistance exoskeleton robot system was designed according to the movement characteristics and torque requirements of different joints.This exoskeleton system simplified the ball-socket structure of the human hip joint by combining the multi-hinged structure and the active-passive joints in the frontal plane and sagittal plane,which satisfied the freedom requirements of the hip joint in these two planes and avoided the movement of the knee joint.The combined structure of linkage and gear mechanism was adopted to simulate the synovial characteristic of the human knee joint and solve the offset problem of the rotation center on the knee joint.An embedded control system based on the bus structure was designed to meet the real-time control requirements of the exoskeleton robot.When each leg of the exoskeleton is in two phases of stance and swing,the wearer has distinctly different requirements for the auxiliary torque and method of the robot.In the stance phase,the exoskeleton is oriented to standing assistance,while in the swing phase,the exoskeleton is oriented to fast-tracking.To satisfy these two distinct requirements,a walking assistance control method based on computational dynamics was studied.The auxiliary supporting method based on position control and the tracking assistance method based on impedance control were established.Aiming at the demands of supporting and fast-swinging during walking,a foot pressure based switching controller was designed to ensure the stability of the system.A virtual potential field based anti-trip control strategy and a safety protection strategy based on human-machine force interaction information were designed to avoid tripping caused by slow reaction speed,impaired judgment and other deterioration of the physically weak persons' body functions during walking.The approach characteristics between feet and obstacles during walking were studied in the anti-trip control strategy,and virtual potential fields were constructed in the direction of forward motion and gravity,which realized the autonomous trip-resistant walking of the wearer.The safety protection strategy blended the advantages on the voluntary movements of the human-dominant mode and the local constraints of the robot-dominant mode.In this strategy,an online trajectory correction method was proposed,which could avoid tripping caused by collision and scraping between feet and obstacles by controlling the distance between them as well as the position of the foothold.To meet the needs of the wearer's active adjustment during the trajectory correction process,a robot active constraint control method with autonomous adaptive ability was designed to realize the blending control of robot trajectory tracking and human movement,which enhanced the safety of the walking process.To ensure the stable switching between the power assisting and trajectory correction,a compliant switching method based on torque and position continuity was proposed to realize the continuous switching of the control system.On the basis of simulation analysis,the joint output ability of the exoskeleton was demonstrated through experiments during walking and climbing stairs,and the effectiveness of the walking assistance algorithm under stance and swing phases was verified.Through the walking experiment in an environment with obstacles,it is verified that the virtual potential field method can effectively improve the ability of the wearer to avoid stumbling during walking.In order to verify the safety assistance strategy based on human-machine force interaction information,an exoskeleton assisted up-stair experiment was designed.The results show that the strategy can enhance the safety of the process.Therefore,the developed exoskeleton robot system can assist the wearer to finish daily activities and walk safely.