Variance constrained control of maneuvering helicopters with sensor failure

This article presents novel results obtained using variance constrained controllers and maneuvering helicopters when also some helicopter sensors fail. For this purpose complex, control oriented, physics based helicopter models are used. A nonlinear model of the helicopter, which includes blade flexibility, is first linearized around specific maneuvering flight conditions (i.e. level banked turn and helical turn). The resulting linearized models are used for the design of variance constrained controllers (i.e. output variance constrained controllers and input variance constrained controllers). Then, the robustness of the closed loop systems with respect to modeling uncertainties (i.e. flight conditions and helicopter inertial parameters variations) is studied. Next, variance constrained controllers are designed for these maneuvering helicopter models when some helicopter sensors fail. Several sensor failure cases are examined and robustness properties of the closed loop systems with respect to modeling uncertainties are also examined. Limitations of the control design process due to the number and type of failed sensors are investigated as well. Finally, the possibility to adaptively switch between controllers in order to mitigate sensor failure is studied.