Optimal H-infinity Control Design under Model Information Limitations and State Measurement Constraints

TL;DR

This paper introduces a suboptimal control design method for interconnected linear systems with limited state measurement and model information, optimizing the $H_ty$ norm through local iterative minimization and maximization.

Contribution

It develops a novel algorithm for control design under partial state measurements and model information constraints, applicable to interconnected systems.

Findings

The algorithm effectively reduces the $H_ty$ norm in control design.

Application demonstrated on vehicle platooning example.

Provides a systematic approach for control under information limitations.

Abstract

We present a suboptimal control design algorithm for a family of continuous-time parameter-dependent linear systems that are composed of interconnected subsystems. We are interested in designing the controller for each subsystem such that it only utilizes partial state measurements (characterized by a directed graph called the control graph) and limited model parameter information (characterized by the design graph). The algorithm is based on successive local minimizations and maximizations (using the subgradients) of the $H_\infty$--norm of the closed-loop transfer function with respect to the controller gains and the system parameters. We use a vehicle platooning example to illustrate the applicability of the results.