Shaping Robust Dynamic Inversion Control of Neural Cell Dynamics

Abstract The human brain consists of 100 billion neurons connected to each other with synapses. When a neuron reaches a threshold voltage, due to the summation of currents flowing in from synapses or by external stimulation, the neuron fires with an action potential, observed as a voltage spike. The nonlinear firing behavior together with the firing rate gives rise to synaptic plasticity and forms the basis of memory, perception, action, and behavior. The perturbed firing of neurons will evolve into undesirable mental states linked to various neurological disorders. Closed-loop control of neuronal firing is thus desirable but challenging due to the underlying highly nonlinear dynamics. In this study, we propose Shaping Robust Dynamic Inversion (SRDI) as a novel robust nonlinear control approach for controlling neuron spiking. We apply SRDI to a Hodgkin-Huxley model to achieve controlled neuron spiking. We compare the performance of SRDI with classical dynamic inversion and linear model predictive control, and our findings demonstrate that SRDI outperforms the other two methods for robust and precise control over firing behavior, enabling neurons to spike at desired moments.