Abstract:System level synthesis enables improved robust MPC formulations by allowing for joint optimization of the nominal trajectory and controller. This paper introduces a tailored algorithm for solving the corresponding disturbance feedback optimization problem for linear time-varying systems. The proposed algorithm iterates between optimizing the controller and the nominal trajectory while converging q-linearly to an optimal solution. We show that the controller optimization can be solved through Riccati recursions leading to a horizon-length, state, and input scalability of $\mathcal{O}(N^2 ( n_x^3 +n_u^3))$ for each iterate. On a numerical example, the proposed algorithm exhibits computational speedups by a factor of up to $10^3$ compared to general-purpose commercial solvers.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to achieve robust model predictive control (RMPC) in linear time - varying systems, especially to optimize the nominal trajectory and the controller through the system - level synthesis (SLS) method. Specifically, the paper proposes a customized algorithm to solve the corresponding disturbance - feedback optimization problem. This algorithm alternately optimizes the controller and the nominal trajectory in each iteration and finally reaches the optimal solution at a q - linear convergence rate. The key contributions of the paper are as follows: 1. **Problem Definition**: The paper considers an uncertain linear time - varying system (LTV system), with the goal of minimizing a cost function while satisfying a series of constraints. The system is affected by external disturbances, which are assumed to be within a unit ball. 2. **Optimization Problem**: Through the SLS method, the robust MPC problem is transformed into a second - order cone programming (SOCP) problem. This SOCP problem includes nominal dynamics, disturbance propagation, and tightening constraints. 3. **Algorithm Design**: The paper proposes an efficient algorithm that solves the above - mentioned SOCP problem by alternately optimizing the nominal trajectory and the controller. The specific steps are as follows: - **Nominal Trajectory Optimization**: Use a QP solver to optimize the nominal trajectory. - **Controller Optimization**: Optimize the controller gain through the Riccati recurrence method. - **Update Dual Variables and Tightening Terms**: Update the dual variables and tightening terms according to the optimization results. 4. **Performance Analysis**: The paper shows the computational efficiency of the proposed algorithm under different state dimensions and prediction horizon lengths through numerical experiments. The experimental results indicate that, compared with general commercial solvers, the proposed algorithm can improve the computational time by up to 10^3 times. 5. **Scalability**: The paper also discusses how to extend this method to nonlinear systems and how to further improve the computational efficiency through parallel computing. In summary, this paper aims to solve the robust MPC problem of linear time - varying systems through the SLS method and proposes an efficient and scalable algorithm to achieve this goal.

Fast System Level Synthesis: Robust Model Predictive Control using Riccati Recursions

Robust Nonlinear Optimal Control via System Level Synthesis

Robust Closed-loop Model Predictive Control via System Level Synthesis

Robust Model Predictive Control of Time-Delay Systems through System Level Synthesis

Robust Optimal Control for Nonlinear Systems with Parametric Uncertainties via System Level Synthesis

Recursive model predictive control for fast varying dynamic systems

Robust Model Predictive Control for nonlinear discrete-time systems using iterative time-varying constraint tightening

A Synthesis Approach of On-Line Constrained Robust Model Predictive Control

Robust model predictive control with polytopic model uncertainty through System Level Synthesis

A Robust Model Predictive Control Scheme for Constrained Linear Systems

Real-time Nonlinear Model Predictive Control using One-step Optimizations and Reachable Sets

Robust Output-Lifted Learning Model Predictive Control

Nonlinear Curve Fitting-Based Fast Robust MPC Algorithm for Nonlinear System

Constructive Robust Model Predictive Control for Constrained Non-Linear Systems with Disturbances

Robust Self-Triggered MPC with Fast Convergence for Constrained Linear Systems

Adaptive Economic Model Predictive Control: Performance Guarantees for Nonlinear Systems

Iterative Convex Optimization for Safety-Critical Model Predictive Control

Robust Explicit Solution of Multirate Predictive Control System with External Disturbances

Robust Mpc for Linear Systems with Structured Time-Varying Uncertainties and Saturating Actuator

Robust model predictive control exploiting monotonicity properties