Loading...

Development of a Biomechanical Model of the Exoped Robot with a User and a Cane

Nasirian, Negin | 2025

85 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 58449 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Behzadipour, Saeed
  7. Abstract:
  8. In this study, a forward dynamics model of the human–lower limb exoskeleton system (ExoPed model) incorporating an instrumented cane was developed to enable simulation of user–system biomechanical interactions and comparison with experimentally measured data. First, a cane equipped with a uniaxial load cell was designed and implemented, capable of measuring the axial force in real time and transmitting it wirelessly to a computer. Subsequently, motion data of the user wearing the exoskeleton and using the cane were collected through a precise marker-based protocol using a three-dimensional motion capture system during walking trials. A full-body dynamic model, including the upper body, lower body, and exoskeleton, was constructed, and inverse kinematics and dynamics analyses were applied to reconstruct joint angles, internal forces, and torques. To evaluate control strategies, a reinforcement learning–based controller was designed and implemented in the forward dynamics model. Model performance was assessed against experimental measurements, showing that the root-mean-square (RMS) error of the user’s motion in the mediolateral and vertical directions within one gait cycle was 6% and 35%, respectively, while the RMS error of the vertical force and torque transferred from the upper to the lower body was 15.8% and 33%, respectively. Furthermore, the mean error of the vertical ground reaction forces between the model and experimental data was 14.34% for the left foot and 15.86% for the right foot. These results demonstrate that the proposed model can accurately reproduce motion patterns and loading conditions, while the implemented control strategies improve stability and energy efficiency. Based on this validated model, the problem of optimal exoskeleton design—including gait parameter tuning, structural mechanics optimization, and actuator controller design—can be pursued more effectively, with higher speed and accuracy, toward objectives such as reducing energy expenditure of both the robot and the user, enhancing user comfort, and improving walking speed
  9. Keywords:
  10. Exoskeleton ; Biomechanical Modeling ; Inverse Kinematics ; Forward Dynamic ; Reinforcement Learning ; Instrumented Cane

 Digital Object List

 Bookmark

No TOC