Tutorial CICA-T Computing with intelligence for identification and control of nonlinear systems

System characterization and identification are fundamental problems in systems theory and play a major role in the design of controllers. System identification and nonlinear control has been proposed and implemented using intelligent systems such as neural networks, fuzzy logic, reinforcement learning, artificial immune system and many others using inverse models, direct/indirect adaptive, or cloning a linear controller. Adaptive Critic Designs (ACDs) are neural networks capable of optimization over time under conditions of noise and uncertainty. The ACD technique develops optimal control laws using two networks - critic and action. There are merits for each approach adopted will be presented. The primary aim of this tutorial is to provide control and system engineers/researchers from industry/academia, new to the field of computational intelligence with the fundamentals required to benefit from and contribute to the rapidly growing field of computational intelligence and its real world applications, including identification and control of power and energy systems, unmanned vehicle navigation, signal and image processing, and evolvable and adaptive hardware systems.