Shaping robust dynamic inversion control of neural cell dynamics.

Area of Science:

• Computational Neuroscience
• Nonlinear Control Systems
• Biophysics

Background:

• Controlling neuronal electrical activity is crucial for understanding brain function and developing neural prosthetics.
• Existing methods struggle with model uncertainty and synaptic noise, limiting precise control over neuron spiking.

Purpose of the Study:

• To develop a robust nonlinear control technique for enforcing precise membrane potential spiking in single neurons and small networks.
• To address challenges posed by model uncertainty and synaptic noise in neural control.

Main Methods:

• Propose Shaping Robust Dynamic Inversion (SRDI), a robust nonlinear control technique.
• Utilize dynamic inversion of neuronal dynamical systems combined with error surface shaping.
• Derive a current control signal to enforce membrane potential spiking.

Main Results:

• SRDI successfully achieved controlled neuron spiking in Hodgkin-Huxley, integrate-and-fire, and FitzHugh-Nagumo models.
• SRDI demonstrated superior robustness compared to classical dynamic inversion.
• SRDI exhibited faster computational time than linear model predictive control.

Conclusions:

• SRDI provides precise and efficient neural control by effectively managing nonlinearities, noise, and model uncertainty.
• The technique enables controlled timing for single spikes, spike trains, and small neuronal networks.
• SRDI represents a significant advancement in robust nonlinear control for neuroscience applications.