Low‐Infrastructure Real‐Time Embedded Control via Variational Integrators

Abstract This paper presents a discrete time receding horizon control scheme that leverages the numerical properties of a variational integrator to facilitate real‐time control generation on an embedded system. The variational integrator employed is well‐suited to classical estimation and control algorithms, e.g. LQR, extended Kalman filters, and particle filters. The structure‐preserving properties of this variational integrator lead to increased performance of estimation and control routines, especially in low‐bandwidth applications. Several experimental examples are presented that illustrate the features of this receding horizon control scheme when leveraging the desirable numerical properties of the variational integrator. © 2016 Wiley‐VCH Verlag GmbH & Co. KGaA, Weinheim)