This Small Business Innovative Research Phase I project will determine computational feasibility of a novel biologically grounded model of supervisory control. The model, called virtual neuronal network, is based on the available data on supervisor cognitive behavior in complex systems, and applies recent theories of neuronal processes to account for the differences in expert and novice performance. The objective of the project is to develop a neuro-controller capable of approximating expert performance in multi-objective, time-constrained tasks. Though some aspects of the model have been validated earlier in different applications, feasibility of the integrated model still needs to be established. Phase I will develop and integrate parts of the virtual neuronal network, and will investigate network performance in sample control problems of realistic size and complexity. Phase II will design and build an efficient neuro-controller for complex systems.
