Loading...

Design a Model Predictive Control System for Empowerment and Rehabilitation of a Lower Limb Exoskeleton

Farghadani, Sahand | 2020

633 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 53112 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Vosoughi, Gholamreza
  7. Abstract:
  8. With the development of technology in the field of control, biomechanics and robotics, wearable robots have found many applications in the field of rehabilitation and empowerment. In the empowerment phase, due to the interaction of these robots with humans, it is necessary to manage the interaction forces between the robot and humans with the help of appropriate control methods. One of the applications of these robots is when a person wants to carry a heavy load attached to the structure of the robot and the robot should be able to transfer the force caused by this load to the ground. At this time, humans should not be exposed to this load and the relationship between the robot and humans should be controlled in such a way as to create the least interactive force. In the rehabilitation phase, one of the applications of this system is to help patients in performing their gait cycles and strengthening patients with mobility problems.This project considers both empowerment and rehabilitation applications. In the case of empowerment, in this project, the development and simulation of a control system for a lower limb two degrees of freedom robot, in order to reduce the interaction forces, so that the interaction force between humans and robots is minimized and the robot allows human to move with the least possible disturbance. In the rehabilitation mode, the goal is to design an integrated system to control the robot with one degree of knee freedom, a functional electrical stimulation system, and an allocation system of these two methods to control the human-robot model. In this way, with the help of muscles, motor energy consumption is reduced and at the same time, the intensity of muscle fatigue of the user is controlled. In order to achieve these goals, the control method based on the nonlinear model predictive control EPFC has been designed for the mentioned systems and its performance in the simulation environment has been investigated. The results showed that the proposed control system was able to reduce the interaction forces between the robot and humans to below 0.5 N in the empowerment phase. Also, in the rehabilitation phase of the control system, it has succeeded in reducing the power consumption of the hybrid robot by 32% and keeping the muscle fatigue factor above 60%
  9. Keywords:
  10. Interaction Force Control ; Functional Electrical Stimulation ; Exoskeleton ; Nonlinear Predictive Control ; Neuromusculoskeletal Model ; Human Robot Interaction (HRI)

 Digital Object List

 Bookmark