摘要
The coupling between the Lyapunov variables and system matrices makes the problem of mixed H2/H∞ flight tracking controller design non-convex. With the aid of enhanced linear matrix inequality （LMI） approach, the non-convex optimization problem is transformed into convex LMI representations. The proposed coupling is eliminated by introducing slack variables. Moreover, a necessary and sufficient condition is derived for the mixed H2/H∞ flight tracking controller which not only stabilizes the controlled system but also satisfies the mixed H2/H∞ performance index in normal case and fault cases. The new enhanced LMI representations provide additional degrees of freedom to solve the non-convex optimization problem, and reduce the conservativeness of the controller design. Simulation results of the aero-data model in a research environment （ADMIRE） model show the advantages of the enhanced LMI approach.
The coupling between the Lyapunov variables and system matrices makes the problem of mixed H2/H∞ flight tracking controller design non-convex. With the aid of enhanced linear matrix inequality （LMI） approach, the non-convex optimization problem is transformed into convex LMI representations. The proposed coupling is eliminated by introducing slack variables. Moreover, a necessary and sufficient condition is derived for the mixed H2/H∞ flight tracking controller which not only stabilizes the controlled system but also satisfies the mixed H2/H∞ performance index in normal case and fault cases. The new enhanced LMI representations provide additional degrees of freedom to solve the non-convex optimization problem, and reduce the conservativeness of the controller design. Simulation results of the aero-data model in a research environment （ADMIRE） model show the advantages of the enhanced LMI approach.
