Bilinear Koopman-based Integrated Guidance and Control for Strap-Down Missiles

This study focuses on the application of integrated guidance and control (IGC) in solving the interception problem of strap-down missiles. The proposed approach involves utilizing the Koopman operator to describe the evolution of the interception process. By representing the system in the space of observable functions, the design of the IGC system for strap-down missiles can be achieved using conventional linear methods, even in the presence of significant nonlinearities. To construct the IGC system, the Koopman operator is approximated through data-driven methods, allowing for the development of both linear and bilinear models. These models are compared, and it is demonstrated that they can effectively capture the strong nonlinear properties of the system, outperforming the traditional local linearized model in terms of accuracy. Furthermore, the bilinear model is argued to be superior to the linear model. By incorporating the bilinear Koopman-based model into a model predictive control framework, the IGC problem with fieldof-view and control constraints can be successfully addressed. Simulations conducted on the interception of strap-down missiles validate the effectiveness and superiority of the proposed IGC scheme.