Seminar: An artificial CPG inspired by a pond snail
Supervisor: Dr. Sharene Bungay
An artificial CPG inspired by a pond snail
Department of Computer Science
Tuesday, November 28, 2017, 11:00 am Room: EN 2022
A Central Pattern Generator (CPG) is a kind of neural network capable of generating rhythmic patterned outputs. Biological CPGs regulate rhythmic motor behaviors such as breathing, feeding, swimming, and digesting, in a particularly efficient way for animals in nature. Similar mechanisms and patterns are studied and extended by researchers in mathematics and physics, and widely applied in control systems such as robotics. In this study, we explore the electrophysiological fundamentals of action potential generation and the Hodgkin-Huxley mechanism of voltage-gated ion channels. An artificial CPG represented by a system of ordinary differential equations (ODEs) was built to replicate the behavior of the respiratory CPG of a pond snail, Lymnaea stagnalis. The Morris-Lecar model was chosen as the starting point to represent each of the neurons in the 3-neuron network of the Lymnaea stagnalis CPG. This model balances the complexity and compliance of the system, and demonstrates the feasibilityof modelling CPG neural networks with limited experimental data using existing neuron models. Delay was introduced into the ODE system, and the resulting artificial CPG approaches the output patterns of the original CPG better than in the ODE system. In this way, delay is shown to be effective for providing better control of timing, as well as richer behaviors in neural networks.