Neural adaptive pointing control of a moving tank gun with lumped uncertainties based on dynamic simulation

This study focuses on the pointing control problem of a moving tank gun. Model uncertainty and foundation vibration, which may be nonlinear, coupled, or time-varying but bounded, are considered. First, the electrohydraulic servo system of a vertical stabilizer is constructed as a nonlinear dynamic system with lumped uncertainties. Second, a neural adaptive controller is proposed to improve the control performance of the vertical stabilizer. A back-propagation neural network is introduced to compensate for the uncertainties, and its weight and threshold values are self-tuned online. Third, a co-simulation model of the moving tank is established. Dynamic simulation verifies that the proposed controller exhibits better performance than typical controllers. Finally, the influence of hull foundation vibration on the proposed controller is analyzed. The pointing accuracy of a moving tank gun is verified to be controlled effectively by the proposed controller under different driving conditions. This work combines control theory with multi-body dynamics to provide a feasible solution for the pointing control problem of a moving tank with model uncertainty and foundation vibration.