Multivariable Integrator Windup Protection for Aircraft Flight ControlSystem using Model Predictive Controller.

The Joint Strike Fighter (JSF) program is underway to develop prototypes for the next generation fighter aircraft. The prototypes in all likelihood will entail a CTOL and STOVL configuration. Both confirgurations are expected tohave numerous control effectors to satisfy the mission requirements. Integrator wind-up protection (IWP) is particularly critical with STOVL control law design problem because the system is so often operating on limits (thrust and vectoring limits) in powered-lift flight. The same situation exists with current Navy Fighters such as F-18 E/F during operation at high angles-of-attack. Complicating factors are (i) the compensators often have a large number of states and (ii) the control law needs to operate on numerous control effectors. Traditional methods of using single-loop-at-a-time protection by freezing the integrators are impractical when state and output variables are numerous and cross-coupled. we propose her the use of a Model Predictive Control (MPC) approach for the design of supervisory controller to satisfy the constraints on the input and output variables for Robust Multivariable Control (RMC) design. MPC approach modifies the command inputs to the RMC controller in an optimal way to keep all the constraints satisfied. Since the disparity between the commanded inpus and the actual inputs to the aircraft are eliminated, no integrator wind-up problem occurs. Phase I will involve the selection of a suitable aircraft simulation with RMC design and the development and testing of MPC for IWP under different flight conditions. SAIC's System Technology Group and Lockheed-Martin's Tactical Aircraft Systems Division will provide technical support during Phases I and II of the project and commercialization support during Phase III.