Study on NNPID-based adaptive control for electro-optical gyro stabilized platform

The working environment of airborne photo-electric stabilized platform is very atrocious and the platform is a time-varying system which has inherent nonlinearity. So we cannot obtain accurate mathematical model which makes it is difficult to obtain good control effects by traditional control method like conventional PID. Given the neural network's self-learning and self-adaptive characteristics which can adjust PID parameters automatically. This paper proposes a feed-forward neural network PID control method based on improved BP (Back Propagation) algorithm. The simulation results show that the improved algorithm can effectively solve the problems caused by classical PID control. Besides, the improved algorithm has very fast response speed, good real-time performance and less static-error. It can indeed meet the practical application requirements.