Abstract:The integration of machine learning methods and Model Predictive Control (MPC) has received increasing attention in recent years. In general, learning-based predictive control (LPC) is promising to build data-driven models and solve the online optimization problem with lower computational costs. However, the robustness of LPC is difficult to be guaranteed since there will be uncertainties due to function approximation used in machine learning algorithms. In this paper, a novel robust learning-based predictive control (r-LPC) scheme is proposed for constrained nonlinear systems with unknown dynamics. In r-LPC, the Koopman operator is used to form a global linear representation of the unknown dynamics, and an incremental actor-critic algorithm is presented for receding horizon optimization. To realize the satisfaction of system constraints, soft logarithmic barrier functions are designed within the learning predictive framework. The recursive feasibility and stability of the closed-loop system are discussed under the convergence arguments of the approximation algorithms adopted. Also, the robustness property of r-LPC is analyzed theoretically by taking into consideration the existence of perturbations on the controller due to possible approximation errors. Simulation results with the proposed learning control approach for the data-driven regulation of a Van der Pol oscillator system have been reported, including the comparisons with a classic MPC and an infinite-horizon Dual Heuristic Programming (DHP) algorithm. The results show that the r-LPC significantly outperforms the DHP algorithm in terms of control performance and can be comparative to the MPC in terms of regulating control as well as energy consumption. Moreover, its average computational cost is much smaller than that with the MPC in the adopted environment.

LPV-Based Offline Model Predictive Control for Free-Floating Space Robots

Optimal Velocity Control of Spherical Robots Based on Offset-free Linear Model Predictive Control

Nonlinear MPC Using an Identified LPV Model

A Model Predictive Control Approach for USV Autonomous Cruising Via Disturbance Learning

Nonlinear Model Predictive Control of Space Redundant Manipulators

Deep Koopman-Operator-Based Model Predictive Control for Free-Floating Space Robots with Disturbance Observer

Model Predictive Control of Constrained LPV Systems.

LPV Based Robust MPC for a Multi-Vectored Airship

Short-Sighted Robust LPV Model Predictive Control: Application to Semi-Active Suspension Systems

Scheduled offline model predictive control based on multiple LPV models

Stabilizing Model Predictive Control for LPV Systems Subject to Constraints with Parameter-Dependent Control Law

Gain-Scheduling Model Predictive Control for Unmanned Airship with Lpv System Description

MPC with One Free Control Action for Constrained LPV Systems

Robust Learning-based Predictive Control for Constrained Nonlinear Systems.

Fast Zonotope-Tube-based LPV-MPC for Autonomous Vehicles

Autonomous Wheel Loader Trajectory Tracking Control Using LPV-MPC

Tube-Based Output Feedback Robust MPC for LPV Systems with Scaled Terminal Constraint Sets.

Dynamic Output Feedback Robust MPC with One Free Control Move for LPV Model with Bounded Disturbance

Robust Learning-based Model Predictive Control for Intelligent Vehicles with Unknown Dynamics and Unbounded Disturbances

Robust Model Predictive Controller Using Recurrent Neural Networks for Input–Output Linear Parameter Varying Systems

Dynamic Output Feedback Robust MPC with Relaxed Constraint Handling for LPV System with Bounded Disturbance