Nonlinear autopilot design for endoatmospheric interceptors with dual control systems

Abstract

acceleration a z is ofa non-minimum phase [3]. By analyzing its phase portrait, the zero dynamics of a tail-controlled missile has been proved to be unstable if the missile's divert acceleration is selected as an output [8,9]. The unstable zero dynamics can also be proved by using the Lyapunov second method [10]. It is well known that a nonlinear control system with an unstable zero dynamics is called a nonlinear non-minimum phase system. Consequently, the non-minimum phase has become an obstacle for applying the feedback linearization technique to the missile's autopilot design. Some approximate linearization methods [8-10] have been applied to overcome the non-minimum phenomenon. It is known that choosing the angle of attack as an output can avoid the non-minimum phase problem, but it indicates that the acceleration guidance command must be transformed into an angle of attack command and such a transformation must cause some error.For a missile with dual control systems, since the reaction thrust can also be involved in the measurement of accelerometers, the autopilot design may be more complicated. It is relatively easy to design an autopilot for the dual controlled missiles based on a classical small deviation linearized model [4,5,11,12] or on choosing the angle of attack as an output in the nonlinear system [6]. However, we should try to analyze the exact nonlinear model of dual controlled missiles, especially its internal dynamics while the divert acceleration is involved in outputs. If the internal dynamics are stable, then a straightforward application of the input-output linearization becomes possible. This paper is organized as follows. Section II presents the nonlinear system model of a missile with dual control systems. In section III internal dynamics of missile models with dual control systems are analyzed. In section IV an autopilot is designed on the basis of internal dynamics analysis. Simulation results are provided in section V. 'IIIAbstract-The nonlinear system model of an endoatmospheric missile whose attitude is controlled by tail fins and reaction jets is presented. By choosing the divert accelerations and rotational rates as output variables, internal dynamics of the nonlinear system are derived and are proved to be bounded under a bounded input. The blending principle of the reaction jets and tail fins is addressed to ensure the normal acceleration is mainly maintained by angle of attack or sideslip angle. An autopilot is designed by using the feedback linearization technique. Results of numerical simulation of an autopilot design example show the effectiveness of the proposed blending principle and the autopilot design.

Design of Missile Autopilot with Aerodynamic Lift and Reaction Jets Based on Auto Disturbance Rejection Controller

“…Advanced interceptor missiles require sufficient maneuver capability and fast response to engage highly maneuverable targets [4]. There is a problem for interceptor missiles of maneuvering targets in the medium endoatmospheric (20000-40000 m) [5].…”

“…Above 20km altitude, if it is only used aerodynamic fins, it is impossible to hit the target, because the control effectiveness of the fins is too weak at this altitude. Advanced interceptor missiles 's attitude is usually controlled by blend control systems comprising of aerodynamic lift and Reaction Control Systems (RCS) to create effective aerodynamic divert acceleration [4]. Dynamics of reaction jet actuators may be faster than dynamics of fin actuators; thus, jets may permit faster missile time cost.…”

“…1. Apparently it is easy to understand that the reaction jet can provide additional forces and angle of attack [4].…”