Radial Basis Function Neural Network-based PID model for functional electrical stimulation system control.

Area of Science:

• Neural Engineering
• Biomedical Engineering
• Rehabilitation Technology

Background:

• Functional electrical stimulation (FES) is a key neural engineering technique for activating paralyzed nerves.
• Traditional controllers struggle with the nonlinear dynamics of FES systems.
• Paraplegia requires advanced assistive technologies for motor function restoration.

Purpose of the Study:

• To design and evaluate a novel FES controller for knee joint movement.
• To improve trajectory tracking accuracy in FES-assisted actions.
• To enhance the control capabilities for individuals with paraplegia.

Main Methods:

• Developed a Proportional-Integral-Derivative (PID) controller integrated with a Radial Basis Function (RBF) neural network.
• Tuned the PID controller using the RBF neural network for adaptive learning and nonlinear control.
• Implemented the RBF-PID FES system to control knee joint angle via lower limb muscle stimulation.

Main Results:

• The RBF Neural Network-based PID model demonstrated superior performance in tracking desired knee angle trajectories.
• Compared to the Ziegler-Nichols tuning method, the RBF-PID system exhibited enhanced accuracy.
• The proposed FES system effectively controlled lower limb muscles for functional movement.

Conclusions:

• The RBF Neural Network-based PID controller offers a more reliable and effective approach for FES systems.
• This advanced control strategy significantly improves FES performance in terms of trajectory tracking.
• The developed system holds promise for restoring functional mobility in individuals with paraplegia.