Intelligent Control Algorithms and PC-Based Software Tool Development for Multiple Effector Control Systems
Small Business Information
500 West Cummings Park Suite, 300, Woburn, MA, 01801
Raman K. Mehra/m. Gopinat
AbstractThe main objective of the proposed work is to develop systematic control design procedures and simulation modules for complex uncertain multiple-input multiple-output (MIMO) systems. Scientific Systems proposes to integrate these modules within a highly flexible and user-friendly Intelligent Control Design Tool (ICDT) for Matlab that will include the design of System Identification (SI) and Model Predictive Control (MPC)-based adaptive control algorithms, and intelligent supervisory control modules based on the system response and fuzzy logic. MPC framework will enable the designer to develop efficient control algorithms for both linear and nonlinear models in the case of hard constraints on control inputs, system outputs, and states. The presence of system nonlinearities and uncertainties will be addressed by both the extensions of several existing identification techniques to nonlinear models, and by the development of MPC-based neural network identification and control algorithms. To preserve robustness under large perturbations, uncertainties, and sensor and actuator failures, the suggested control system will have substantial reconfiguration capabilities introduced by supervisory control based on on-line adaptation, fuzzy rules and Fault Detection and Isolation (FDI) mechanisms. The feasibility of the intelligent MPC algorithms will be tested through simulations of Eagle Eye UAV, or other Uninhabited Combat Air Vehicle (UCAV) model recommended by the Navy. Prof. K.S. Narendra of Yale University will provide technical consulting support.
* information listed above is at the time of submission.