State-space multivariable non-linear identification and control of aircraft

Abstract

In the design of commercial and military aircraft, flying and handling qualities, as well as performance (agility and manoeuvrability, controllability and stability) and safety, have received increased emphasis. The aircraft must exhibit acceptable handling qualities while performing manoeuvres, and different flight management systems have been designed. For advanced aircraft, the assumption that longitudinal and lateral dynamics can be decoupled leads to poor flying and handling characteristics owing to longitudinal-lateral cross-coupling. Aircraft stability severely degraded at low speed, high-angle-of-attack flight and gain-scheduled linear control algorithms cannot provide the desired performance (gain-scheduled control laws are used because they are certifiable). These have resulted in the critical need for significant improvements in flying and handling qualities as well as expansion of the flight envelope at low-speed, high-angle-of-attack flight by using robust non-linear controllers. Military aircraft must be capable of performing a diverse range of operational missions, which include air-to-ground and air-to-air combat. Mission effectiveness relies on flight management systems and control algorithms. Control system effectiveness can be achieved as the designer integrates non-linear flight dynamics and advanced optimization methods to design multivariable control algorithms. In this paper a study is made of non-linear identification and control methods which can be applied to multivariable aircraft using measured variables. By performing non-linear modelling, it is shown that the methods reported lead to very encouraging results.