Research on attack angle tracking of high speed vehicle based on PID and FLNN neural network

Abstract

A kind of pitch channel dynamic model of hypersonic aircraft considered with both the model of engine and the model of elastic shape is studied. A special typical flying point is chosen that the state of elastic shape is assumed to be constant and the engine is designed with a PID law to make the speed of aircraft close to a constant. Then a kind of hybrid controller based on FLNN neural network and PID control is designed to make the attack angle can converged to desired value. And the adoption of Taylor type FLNN neural network can make use of the form of air dynamic coefficient which is a function like Taylor series, so it has a good adaptive ability to compensate the uncertainties and unconsidered factors of hypersonic model. And the use of PID control law can make use of the advantage of traditional classic control theory and what is the most important of all is that the PID control can provide enough damp ratio to make the system stable enough. So the hybrid control strategy can integrate both advantage of neural network and PID control methods which is also testified by the detailed numerical simulation in the last part of this paper.