Immersion- and invariance-based adaptive missile control using filtered signals

Abstract

The article presents a new non-certainty-equivalent adaptive (NCEA) longitudinal autopilot for the control of a missile based on the immersion and invariance theory. The interest here is to control the angle of attack of the missile in the presence of large parametric uncertainties. For the derivation of the control law, a backstepping design procedure is used. At each step of the design, certain filtered signals are generated for the synthesis of a stabilizing control signal and a parameter estimator. Using Lyapunov stability analysis, it is shown that in the closed-loop system, trajectory control of the angle of attack is accomplished, and the trajectories of the system are attracted to certain manifold in the space of state variables and parameter errors. For stability in the closed-loop system, an explicit analytical relation involving the controller gains is obtained. It may be pointed out that recently an adaptive autopilot based on the immersion and inversion theory has been designed, but it has stringent requirements because for its synthesis, the derivatives of the Mach number and angle of attack must be known, and a large number of parameters must be updated. The derived control system of this article is synthesized using only the state variables, and its identifier is of lower order. A traditional certainty-equivalent adaptive autopilot is also presented for comparison. Simulation results are obtained which show that the designed NCEA control system can accomplish angle of attack control despite large parametric uncertainties; and it can give better tracking performance than the traditional controller.