In cooperation with industrial partner Amerigon, Inc., NeuroDyne, Inc., plans a new thrust toward electric and hybrid vehicles. Previous work by NeuroDyne has developed methods of identification and intelligent control, using artificial neural networks (ANN) which are well adapted to work with vehicle systems. Part of this effort (supported by NSF Grant ECS 9216530 `Reducing Pollution and Increasing Fuel Economy Using Intelligent Control` through the University of South Carolina) resulted in a rational scheme for identifying, with feedforward neural networks, a class of nonlinear dynamic systems. The major problem to be addressed is to apply the NeuroDyne identification methods to an electric energy management system (EMS); the goal of the EMS is to optimize driving range by budgeting power demands, while taking into account safety, driveability and passenger comfort. An extension of this work is to provide ANN-based controls for hybrid automobiles (fuel-burning engine plus batteries) taking into account fuel economy, exhaust emissions, and efficiency.
