Gain-Scheduled Autopilot Design with Anti-Windup Compensator for a Dual-Spin Canard-Guided Projectile

Abstract

This paper explores the design of an autopilot for a dual-spin guided projectile taking into account input saturations. The projectile full nonlinear model is described and then put in a quasi-LPV form suitable for controller synthesis. A baseline gain-scheduled autopilot is then computed without taking into account the saturation nonlinearities. As a major contribution of this paper, the impact of saturations is next highlighted through a degraded flight scenario, and an antiwindup compensator is added to the closed-loop to counteract their effects. The improvements brought by the anti-windup are then evaluated through Monte Carlo simulations.

An Interpolated Model Recovery Anti-Windup for a Canard-Guided Projectile Subject to Uncertainties

“…The baseline autopilot considered in this paper is identical to the one used in [2], and only its main features are recalled here. The controller is structured as in Fig.…”

“…5. In [2], local static anti-windup compensators were computed and then interpolated to cover the flight envelope. While the simplicity of this structure makes it attractive, it is interesting to investigate whether we can benefit from the increased number of degrees of freedom of a dynamic anti-windup.…”

“…Previous work on this concept include [1], which tackled system modelling as well as controller synthesis. More recently in [2], saturations were taken into account and a gain-scheduled static anti-windup compensator was proposed. Building up on these previous studies, this paper revisits the anti-windup design and proposes a dynamic compensator with a suitable interpolation strategy, and sets up the integral quadratic constraints (IQC) framework for robustness analysis of the closed-loop against both aerodynamic uncertainties and saturations.…”

“…French Aerospace Lab, Toulouse, France <clement.roos, jean-marc.biannic>@onera.fr Section II reviews the open-loop components and model. Section III deals with autopilot design, with a brief description of the baseline autopilot from [2], and the computation of a dynamic anti-windup compensator. Section IV presents the robustness analysis framework and early analysis results, and Section V details Monte Carlo simulations.…”

L1 Adaptive Augmentation of an Incremental Nonlinear Dynamic Inversion Autopilot for Dual-Spin Guided Projectiles

“…Moreover, the lateral dynamics are strongly coupled due to the high spin rate of the main body of the projectile. All these elements make the control of the projectile difficult by the means of classical and linear control techniques like gain-scheduling [2,3] because linear controllers designed for numerous operating points are needed to cover the flight envelope. Besides, gain-scheduling can guarantee performance and stability only locally around these flight points [4].…”