Research On Pilot-aircraft System Controller Design Based On Unfalsified Control Theory

Civil flight safety is a world-concerning question.Flight crew human factors has a great influence on flight safety,meanwhile the human factor analysis has important meaning for aircraft design.Pilot model for flight safety quantification is a new pilot model,which represents the pilot-aircraft interface and the characteristics of pilot in flight activity.For the pilot-aircraft system with such pilot model,no controller design is introduced to meet the time-varying characteristics caused by human factors.Unfalsified adaptive control is a model-free adaptive data-driven control theory.Without detailed description about model structure and parameters,it delivers control with the measured input/output data of the system.The key point of this thesis is research on pilot-aircraft system controller design based on unfalsified control.This thesis mainly includes the following aspects: building a six-degree of freedom aircraft model,introducing the pilot model and designing a longitudinal control system;introducing basic concept of unfalsified adaptive control,proposing a controller design method based on classical unfalsified adaptive control method,analyzing the stability,and verifying the validity with simulation method;considering the potential bad effect of the old data on controller switching,proposing another controller design method based on fading memory unfalsified adaptive control method and simulation;At last analyzing the effect of different fading memory factors have on flight quality.Research result indicates that the two control method proposed can guarantee the stability of the pilot-aircraft system.The control method controls system with system input/output data and switches when necessary.The methods decrease the flight trajectory error effectively.