A novel active disturbance rejection-based control strategy for a gun control system

To compensate for the nonlinearity and to achieve finely-tuned tracking accuracy of a gun control system driven by an AC machine, an improved active disturbance rejection control (IADRC) strategy with neural network embedding (NN-IADRC) is developed in this paper. The proposed IADRC, which has amnestic memory effects, can be regarded as an extension of the conventional ADRC (CADRC), making it a special case of the IADRC. To further attenuate the dependence on system models and enhance the disturbance rejection capacities of the IADRC strategy, an on-line NN-based optimum updating approach is also developed in this paper. Finally, a series of experiments are conducted on the semi-physical simulation platform to estimate the performance of the control system and the effects of the memory factor on the system. The experimental results confirm that the proposed NN-IADRC is highly robust. The results also confirm that it performs more excellently than the CADRC and that its fine tuning has attained tracking accuracy.