Simulation of two-rate adaptive neural network and fuzzy logic hybrid control for stochastic model of an experimental aircraft

摘要

The nature of the multirate dynamics of a process makes it very attractive for applications, since the multirate phenomena are complex. This paper presents a research methodology for describing a two-rate stochastic control system as state-space (SS) type decomposed models of multi-input/multi-output (MIMO) stochastic control subsystems with "fast" and "slow" adaptive neural networks (NNs), and with neuro-fuzzy networks (NFNs) and fuzzy logic (FL) control structures. The block diagrams for both the original system with a linear-quadratic-Gaussian (LQG) regulator and the decomposed subsystems with two-rate adaptive NNs and FL hybrid control for the stochastic model of a tracking system for an experimental aircraft were designed. The simulation results demonstrate that this research technique would work for real-time MIMO stochastic systems.