Incremental Nonlinear Dynamic Inversion Autopilot with Adaptive Augmentation for Guided Projectiles

Abstract

This paper covers the design of an L1-adaptive incremental nonlinear dynamic inversion (INDI) autopilot applied to the correction of the ballistic dispersion of a 155 mm dual-spin projectile equipped with a rolldecoupled course-correction fuze. First, an INDI autopilot baseline is designed with a tuning methodology taking into account some implementation constraints (i.e., actuator bandwidth and sampling frequency). This paper highlights the degrading effect of these constraints on the autopilot performance through INDI inner-loop analysis. Then, an adaptive augmentation scheme is presented to dynamically compensate for the degraded model inversion in the INDI autopilot due to parametric uncertainties. Monte Carlo simulations for trajectory correction scenarios are performed on the uncertain model. Performance comparison between baseline and augmented autopilot highlights the benefits of implementing this adaptive scheme in terms of both robustness to parametric uncertainties and reduction of dispersion.