Co-Design of Lipschitz Nonlinear Systems

TL;DR

This paper introduces a novel co-design approach for Lipschitz nonlinear systems, optimizing plant and controller simultaneously to improve performance and resource efficiency, with a proven stable solution method demonstrated on a robot example.

Contribution

It presents a new time-independent reformulation of the co-design problem and a gradient-based solution procedure ensuring system stability.

Findings

Effective co-design method for Lipschitz nonlinear systems.

Stable solutions verified through a robot example.

Improved performance and resource savings demonstrated.

Abstract

Empirical experiences have shown that simultaneous (rather than conventional sequential) plant and controller design procedure leads to an improvement in performance and saving of plant resources. Such a simultaneous synthesis procedure is called as "co-design". In this letter we study the co-design problem for a class of Lipschitz nonlinear dynamical systems having a quadratic control objective and state-feedback controller. We propose a novel time independent reformulation of the co-design optimization problem whose constraints ensure stability of the system. We also present a gradient-based co-design solution procedure which involves system coordinate transformation and whose output is provably stable solution for the original system. We show the efficacy of the solution procedure through co-design of a single-link robot.