Research On Gait Teaching Of Lower Limb Exoskeleton Based On Inertial Sensor Dat

The research objective of this study was to design a gait training method that possessed the advantages of convenience,flexibility,ease of operation,low cost,and high real-time performance,aiming to provide long-term professional gait rehabilitation therapy for communities and families.Based on the above objectives,this study proposes a gait teaching method for lower limb exoskeleton robots by utilizing data from individual inertial sensors placed on the dorsal part of the feet.The rotational angles of the three lower limb joints are calculated in real-time through coordinate transformation and inverse kinematics algorithms,enabling the real-time implementation of gait teaching for lower limb exoskeleton robots.To verify the feasibility of this method and the consistency of the teaching trajectory,this paper conducted a comparative analysis between the measured joint angle data from three inertial sensors worn on the lower limb joints of the teaching personnel and the exoskeleton angles.Inertial motion capture devices were used to obtain the data for analysis.The main research work of this paper is as follows:(1)A gait teaching algorithm based on a single inertial sensor was proposed,and the feasibility of the algorithm was verified using the measured joint angle data obtained from three inertial sensors placed on the lower limb joints using motion capture devices.The paper elucidated the process of coordinate transformation for the inertial sensor data.The lower limb exoskeleton model was established based on the Denavit-Hartenberg(D-H)parameters method.Analysis and derivation of the inverse kinematics equations were conducted to obtain the joint angle data for the three lower limb joints.The algorithm’s feasibility was validated by controlling the lower limb exoskeleton to perform teaching movements and verifying the results.(2)The lower limb exoskeleton was modeled in 3D using Solid Works.In terms of structure,the knee and ankle joint driving motors were relocated upward to align with the superior drive axis,reducing the influence of motor inertia and ensuring smooth joint movement.Simulink was used to conduct dynamic system simulations of the lower limb exoskeleton to validate the reliability of the power system.Through the verification process,it was determined that the selected motors met the required driving specifications.(3)Based on the aforementioned simulation results,a physical prototype of an exoskeleton leg was fabricated,comprising hip,knee,and ankle joints.Subsequently,a measurement system was established,consisting of a real teaching subject,an inertial sensor strapped to the dorsal surface of the subject’s foot,the physical prototype,and a control system.Utilizing a control algorithm developed in this study,a relatively complete human-machine teaching experimental system was realized.To validate the practical applicability of the algorithm proposed in this paper,a comparative analysis was conducted between the measured joint angle data obtained from the inertial sensors placed on the lower limb joints of the teaching subject using the inertial motion capture system and the corresponding exoskeleton joint angle data.Through the comparative analysis,it was determined that there was a delay of within 1 second in the joint angle measurements between the prototype and the inertial sensors.The angular error was within 2°,thus meeting the requirements for gait rehabilitation training.This real-time analysis validated the real-time nature and accuracy of the gait teaching algorithm based on a single inertial sensor,establishing a foundation for providing more scientifically grounded gait training for patients in community or home settings.