Disturbance Accommodating Controller for Uncertain Stochastic Systems with Controller Saturation Jemin George, * Puneet Singla, † and John L. Crassidis ‡ University at Buffalo, State University of New York, Amherst, NY, 14260-4400 Observer based stochastic disturbance accommodating control scheme utilizes an es- timator to determine the necessary corrections to the nominal control input and thus minimizes the adverse effects of both model uncertainties and external disturbances on the controlled system. The total control input, which includes the nominal control as well as the control corrections to the nominal control to accommodate for the disturbances, could exceed the actuator saturation limits. The saturation of the disturbance accommodating control system is more violent than that of the nominal control system because of the positive feedback of the disturbance compensation. This paper presents the formulation of the stochastic disturbance accommodating controller with actuator saturation for a class of uncertain linear stochastic systems. A Lyapunov based stochastic adaptive approach is used to update the control gains and the process noise covariance online so that the closed-loop stability of the controlled system is guaranteed. I. Introduction External disturbances and system uncertainties can obscure the development of a stable control law. The main objective of disturbance accommodating controller is to make necessary corrections to the nominal con- trol input to accommodate for external disturbances and system uncertainties. 1–4 The disturbance accom- modating observer approach has shown to be extremely effective for disturbance attenuation. 5–7 However, due to observer gain sensitivity to the external disturbances, performance of the observer can significantly vary for different types of exogenous disturbances. In our previous work, 8, 9 a robust control approach based on a significant extension of the observer based disturbance accommodating control concept is presented. The robust control approach compensates for model parameter uncertainties and external disturbances by estimating a model-error vector in real time that is used as a signal synthesis adaptive correction to the nominal control input to achieve maximum performance. Utilizing a Kalman filter in the feedback loop, this control approach simultaneously estimates the system states and the model-error vector or the disturbance term from noisy measurements. 10–13 The estimated states are then used to develop a nominal control law while the estimated disturbance term is used to make necessary corrections to the nominal control input to minimize the adverse effects of system uncertainties and the external disturbances. One of the major disadvantages of the stochastic disturbance accommodating controller is that the total control input, which includes the nominal control as well as the control corrections to the nominal control to accommodate for the disturbances, could exceed the actuator saturation limits. When the desired input exceeds the actuator limits, the controller fails to accommodate for the disturbances and this could drive the system unstable. Dealing with actuator saturation has been well recognized to be practically imperative yet a theoretically challenging problem. When a control system has integral compensators, the control saturation could lead to an unstable response, which is called the integral wind-up phenomenon. There is very little research about the wind-up phenomenon for disturbance accommodating observer based controllers. The wind-up phenomenon of the disturbance observer system is more vulnerable than that of the integral control system because of the positive feedback of the disturbance compensation. A wind- up restraint disturbance accommodating controller using a saturation model in the disturbance observer is * Ph.D. Student, Department of Mechanical & Aerospace Engineering, jgeorge3@buffalo.edu, Student Member AIAA. † Assistant Professor, Department of Mechanical & Aerospace Engineering, psingla@eng.buffalo.edu, Member AIAA. ‡ Professor, Department of Mechanical & Aerospace Engineering, johnc@eng.buffalo.edu, Associate Fellow AIAA. 1 of 16 American Institute of Aeronautics and Astronautics