Digital System Identification Controller for Adaptive Feedback Control in Closed-Loop FES

Abstract

Functional Electrical Stimulation (FES) devices are widely used for spinal cord injury patients but often face challenges with nonlinearity effects, leading to premature muscle fatigue due to feedback controller discrepancies. Therefore, this research proposes a digital system identification controller (SIC) to determine the patient's condition by extracting important information from the patient’s knee trajectory response, which includes time delay, rise time, overshoot, steady-state time, steady-state value, and steady-state error. The extracted information is beneficial in enabling real-time self-tuning of the feedback control system based on the patient’s condition. Initially, the digital SIC was modelled using MATLAB Simulink, then hardcoded into digital logic using hardware description language (HDL) Verilog codes and verified using Quartus Modelsim. Thereafter, the performance of the digital SIC was tested with a knee extension model using the HDL co-simulation technique in MATLAB Simulink. The results showed that the digital SIC designed in HDL produces accurate output reading as MATLAB Simulink with error percentages ranging from 0% to 33.3%.  The adaptive feedback controller could use this key data information to fine-tune the internal feedback settings. This evaluation demonstrates the controller's functionality and efficiency, providing insights into its suitability for adaptive feedback control applications.