Development and Analysis of a Nonlinear Dynamic Inverse Control Strategy
MetadataShow full item record
Aircraft control normally relies on gain-scheduling of linear control laws, an approach that has been very successful. One can characterize nonlinear dynamics in high-alpha maneuvers by using a large number of linear models, but his method may not be adequate, especially since some of the chosen operating points may not be equilibrium points. One can build robustness into the control strategy (e.g., by requiring more gain and phase margins), but this approach translates to a more conservative control, which means possible performance degradation. An alternative method is to take explicitly into account the nonlinearities in the control design, thus better utilizing the existing dynamics and control power. We choose to investigate the dynamic inverse control technique because of its ease of implementation, and the simply way that maneuvers can enter into the control. The objective of nonlinear dynamic inversion is to invert the dynamic equations of the plant directly in order to find the control necessary to yield the given output.<P>We elaborate the dynamic inverse methods first described by Meyer and Cicolani. We expand the method to a more complex aerodynamics and airframe description, that of the full nonlinear simulation (wind-tunnel and flight tested model) of the X-29. To achieve additional realism, the simulation contains actuator redundancy and actuator limits. We first formulate a reduced analytic model in order to use feedback linearization techniques to design the controller. Since we neglect some of the aerodynamic terms, the controller is then modified so that stability robustness to modeling errors can be achieved. In addition, we modify the robust control method to add integral action to enable the controller to reduce steady state errors and to lower the control rates.