There are two main approaches to design autopilot and guidance systems: “Single-Loop” and “Two-Loop”. In the “Single-Loop” approach there is no definite autopilot, and the guidance command is applied directly to the actuators. In the “Two-Loop” approach the inner autopilot loop has to follow commands generated by the outer guidance loop, and each loop is designed separately. The main goal of the research was to identify the merits of each approach, and combine the benefits of the two approaches into a new improved one.
In the talk, the advantages of the “Two-Loop” and “Single-Loop” approaches will be discussed. The conditions for obtaining equivalence between the two design approaches will be presented. Broadly speaking, an equivalence relation between these two design approaches holds if full-state feedback exists and the number of guidance commands is equal to the number of available controls. The equivalence conditions are established under linear quadratic optimal control and differential game formulations, and for nonlinear systems with bounded controls. The performance of the “Two-Loop” and “Single-Loop” approaches to design autopilot and guidance systems is illustrated via simulations.