Robust self-triggered DMPC for linear discrete-time systems with local and global constraints

TL;DR

This paper introduces a robust self-triggered distributed model predictive control scheme for linear discrete-time systems with local and global constraints, utilizing tube-based methods and ADMM for efficient, parallel optimization.

Contribution

It presents a novel self-triggered DMPC approach combining tube-based disturbance handling and ADMM-based distributed optimization for systems with coupled constraints.

Findings

Ensures recursive feasibility and stability of the control system.

Reduces computational load through self-triggering mechanism.

Demonstrates effectiveness via simulation example.

Abstract

This paper proposes a robust self-triggered distributed model predictive control (DMPC) scheme for a family of Discrete-Time linear systems with local (uncoupled) and global (coupled) constraints. To handle the additive disturbance, tube-based method is proposed for the satisfaction of local state and control constraints. Meanwhile, A special form of constraints tightening is given to guarantee the global coupled constraints. The self-triggering mechanism help reduce the computation burden by skip insignificant iteration steps, which determine a certain sampling instants to solve the DMPC optimization problem in parallel ways. The DMPC optimization problem is constructed as a dual form, and solved distributedly based on the Alternative Direction Multiplier Method (ADMM) with some known simplifications. Recursive feasibility and input-to-state stability of the closed-loop system are shown, the performance of proposed scheme is demonstrated by a simulation example.