Design of an H∞ Gain-Scheduled Guidance Scheme for a Guided Projectile

This paper addresses the design of a two-dimensional guidance scheme in the vertical plane, using robust and gain-scheduling control methods, for a fin-stabilized canard-guided projectile. Given that the conventional proportional navigation guidance (PNG) is not completely adequate for target interception with a control-authority-limited weapon, the proposed guidance law is offered as an improvement compared with PNG while respecting a similar parallel navigation guidance principle. The projectile-target engagement kinematics full nonlinear model is discussed, and a linear parameter-varying model, necessary for guidance design, is derived without any approximation. A set of reduced-order mixed-sensitivity H-infinity controllers with an additional reference model following constraint is designed using a specific computation procedure. The resulting homing closed-loop system is assessed through linear and nonlinear simulations of either nominal or perturbed scenarios. These scenarios include noise or uncertain launch initial conditions, and they also take a simplified or a more realistic autopilot loop into account. The gain-scheduled guidance scheme is finally compared with PNG, and with three other recent guidance laws considering limitations on the acceleration command.