Loading...

Design and Implementation of a FES-Assisted Control for a Lower Limb Hybrid Exoskeleton Robot in order to Reduce the Robot Energy Consumption

Karami, Mohammad | 2021

560 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 53851 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Vosoughi, Gholamreza; Moradi, Hamed
  7. Abstract:
  8. Wearable robots, or exoskeletons, are electromechanical devices that aim to assist user by supplying additional mechanical power as needed. The interaction of these robots with humans and the possibility of robots using their various abilities such as intelligence, control power, sensors, etc., make one of the crucial components of these control algorithms, its interaction and integration with the user. In this regard, functional electrical stimulation is one way to use human abilities to control the robot and achieve broader goals. Adding functional electrical stimulation to a wearable robot creates an emerging group of robots called hybrid exoskeletons. Hybrid exoskeletons provide the ability to combine functional electrical stimulation with a wearable robot to better control joints and reduce the patient's metabolic cost.This project aims to design and implement a control method using functional electrical stimulation to reduce the energy consumption of the exoskeleton robot, taking into account the muscle fatigue factor. To conduct this research, first in the theoretical aspect, using a new neuro-musculoskeletal model that can simulate knee torque due to electrical stimulation and the dynamic model of a noble wearable robot, the hybrid robot system in Simulink MATLAB is simulated. The dynamic continuous fractional-order nonsingular terminal sliding mode controller (DCFONTSM) is implemented on this model. Also, to achieve the goal, an intelligent strategy is designed to manage muscle fatigue while following the optimal gate path of the knee. The results obtained from the simulations indicate a maximum tracking error of 0.05° for the DCFONTSM controller. Also, in the hybrid mode, this controller reduces the normalized power consumption of the exoskeleton robot by about 32% compared to the case without applying electrical stimulation.The final step is to implement the control system designed on sharif hybrid exoskeleton. The first test is implementing the TDE-DCFONTSM controller with an intelligent allocation strategy based on the fitness function. The second experiment is the implementation of the TDE-DCFONTSM controller with an allocation strategy based on the EMG signal, the results of which show an improvement in fatigue prediction and a reduction of about 18% RMS of normalized power consumption of the robot in hybrid mode
  9. Keywords:
  10. Lower Limb Exoskeleton ; Spinal Cord Injury ; Rehabilitation ; Functional Electrical Stimulation ; Muscular Fatigue ; Energy Consumption Reduction

 Digital Object List

 Bookmark

No TOC