Research on Neutral Network Based Impact Angle Control Considering the Field-of-view Limitation

The impact angle control guidance problem considering the strapdown seeker’s field-of-view has become an interested topic and it has been solved by various techniques. However, most of the existing solutions suffer from undesirable fluctuation in their guidance commands due to the disturbance of the nonlinear system. In this paper, we design a field-of-view constrained impact angle control guidance law by using an adaptive RBF neural network based sliding mode controller. In the design of the controller, a logarithmic barrier Lyapunov function and a quadratic Lyapunov function are used for forcing the system to reach the sliding mode in limited time and a hyperbolic tangent function is introduced to solve the field-of-view limitation. An adaptive RBF neural network is constructed to approximate the system’s uncertain disturbance and the approximation serves as compensation in the guidance command to mitigate the adverse fluctuation. Finally, the performance of the proposed solution is verified by numerical simulations through two engagement scenarios.