Data-Driven Model Predictive Control with Stability and Robustness Guarantees

Abstract

We propose a robust data-driven model predictive control (MPC) scheme to control linear time-invariant (LTI) systems. The scheme uses an implicit model description based on behavioral systems theory and past measured trajectories. In particular, it does not require any prior identification step, but only an initially measured input-output trajectory as well as an upper bound on the order of the unknown system. First, we prove exponential stability of a nominal data-driven MPC scheme with terminal equality constraints in the case of no measurement noise. For bounded additive output measurement noise, we propose a robust modification of the scheme, including a slack variable with regularization in the cost. We prove that the application of this robust MPC scheme in a multi-step fashion leads to practical exponential stability of the closed loop w.r.t. the noise level. The presented results provide the first (theoretical) analysis of closed-loop properties, resulting from a simple, purely data-driven MPC scheme.

References in corpus (6)

• Cautious Model Predictive Control using Gaussian Process Regression
• Safe Reinforcement Learning Using Robust MPC
• A trajectory-based framework for data-driven system analysis and control
• Robust Constraint Satisfaction in Data-Driven MPC
• Data-Driven Tracking MPC for Changing Setpoints
• From self-tuning regulators to reinforcement learning and back again

Cited by in corpus (89)

• Model Based Control of Soft Robots: A Survey of the State of the Art and Open Challenges
• Linear tracking MPC for nonlinear systems Part II: The data-driven case
• Combining Prior Knowledge and Data for Robust Controller Design
• A trajectory-based framework for data-driven system analysis and control
• Data-Driven Control for Linear Discrete-Time Delay Systems
• DeeP-LCC: Data-EnablEd Predictive Leading Cruise Control in Mixed Traffic Flow
• On Recurrent Neural Networks for learning-based control: recent results and ideas for future developments
• Robust stability analysis of a simple data-driven model predictive control approach
• Maximum Likelihood Estimation in Data-Driven Modeling and Control
• Linear tracking MPC for nonlinear systems Part I: The model-based case
• Event-triggered Consensus Control of Heterogeneous Multi-agent Systems: Model- and Data-based Analysis
• An overview of systems-theoretic guarantees in data-driven model predictive control
• Efficient Off-Policy Q-Learning for Data-Based Discrete-Time LQR Problems
• On a Stochastic Fundamental Lemma and Its Use for Data-Driven Optimal Control
• Data-driven Control of Dynamic Event-triggered Systems with Delays
• Data-driven model predictive control: closed-loop guarantees and experimental results
• Robust Constraint Satisfaction in Data-Driven MPC
• On the design of terminal ingredients for data-driven MPC
• Data-Driven Tracking MPC for Changing Setpoints
• Constrained nonlinear output regulation using model predictive control -- extended version
• Data-driven Dynamic Event-triggered Control
• Data-Driven Gain Scheduling Control of Linear Parameter-Varying Systems using Quadratic Matrix Inequalities
• Data-based Control of Feedback Linearizable Systems
• On a Continuous-Time Version of Willems' Lemma
• Data-based Transfer Stabilization in Linear Systems
• Data-based guarantees of set invariance properties
• Data-driven predictive control with improved performance using segmented trajectories
• Amidst data-driven model reduction and control
• Learning-Based Modeling of Human-Autonomous Vehicle Interaction for Improved Safety in Mixed-Vehicle Platooning Control
• Data-driven stabilization of switched and constrained linear systems
• Near-Optimal Design of Safe Output Feedback Controllers from Noisy Data
• Data-driven synthesis of Robust Invariant Sets and Controllers
• Distributed Data-driven Predictive Control via Dissipative Behavior Synthesis
• Data-driven Dissipativity Analysis of Linear Parameter-Varying Systems
• Towards a Systems Theory of Algorithms
• Uncertainty-aware data-driven predictive control in a stochastic setting
• Data-driven input reconstruction and experimental validation
• On Regularizability and its Application to Online Control of Unstable LTI Systems
• Towards data-driven stochastic predictive control
• Harnessing Uncertainty for a Separation Principle in Direct Data-Driven Predictive Control
• Online convex optimization for data-driven control of dynamical systems
• Enhancing Safety in Mixed Traffic: Learning-Based Modeling and Efficient Control of Autonomous and Human-Driven Vehicles
• Decentralized Data-Enabled Predictive Control for Power System Oscillation Damping
• Stability in data-driven MPC: an inherent robustness perspective
• Practical one-shot data-driven design of fractional-order PID controller: Fictitious reference signal approach
• Data-driven distributed MPC of dynamically coupled linear systems
• On the impact of regularization in data-driven predictive control
• Adaptive Koopman Embedding for Robust Control of Complex Nonlinear Dynamical Systems
• Bayesian model predictive control: Efficient model exploration and regret bounds using posterior sampling
• A quantitative and constructive proof of Willems' Fundamental Lemma and its implications
• Virtual reference feedback tuning with robustness constraints: A swarm intelligence solution
• Direct Data-Driven State-Feedback Control of Linear Parameter-Varying Systems
• On the design of autonomous agents from multiple data sources
• Implicit predictors in regularized data-driven predictive control
• Data-driven control on encrypted data
• Learning control for polynomial systems using sum of squares relaxations
• On the equivalence of direct and indirect data-driven predictive control approaches
• Adaptive Output Feedback Model Predictive Control
• Data-Driven Prediction with Stochastic Data: Confidence Regions and Minimum Mean-Squared Error Estimates
• A System Parameterization for Direct Data-Driven Estimator Synthesis
• Data-Driven Resilient Predictive Control under Denial-of-Service
• An Online Adaptation Strategy for Direct Data-driven Control
• On Controllability and Persistency of Excitation in Data-Driven Control: Extensions of Willems' Fundamental Lemma
• Physically Consistent Multiple-Step Data-Driven Predictions Using Physics-based Filters
• Robust Data-Driven Tube-Based Zonotopic Predictive Control with Closed-Loop Guarantees
• Direct data-driven control of LTV systems
• Stability and Performance Analysis of Model Predictive Control of Uncertain Linear Systems
• Frequency-Domain Data-Driven Predictive Control
• Adaptive Stochastic Predictive Control from Noisy Data: A Sampling-based Approach
• Online Optimization and Ambiguity-based Learning of Distributionally Uncertain Dynamic Systems
• Non-conservative Design of Robust Tracking Controllers Based on Input-output Data
• Data-driven observer design for an inertia wheel pendulum with static friction
• Sampling-based Stochastic Data-driven Predictive Control under Data Uncertainty - Extended Version
• Data-Driven Reachability Analysis Using Matrix Zonotopes
• Adaptive Output Feedback MPC with Guaranteed Stability and Robustness
• Extending direct data-driven predictive control towards systems with finite control sets
• Behavioural Approach to Distributed Control of Interconnected Systems
• Adaptive FRIT-based Recursive Robust Controller Design Using Forgetting Factors
• Offline Uncertainty Sampling in Data-driven Stochastic MPC
• On data usage and predictive behavior of data-driven predictive control with 1-norm regularization
• Comparative Analysis of Data-Driven Predictive Control Strategies
• Online Stochastic Optimization for Unknown Linear Systems: Data-Driven Synthesis and Controller Analysis
• Control of Unknown (Linear) Systems with Receding Horizon Learning
• Confidential FRIT via Homomorphic Encryption
• Stochastic Data-driven Predictive Control of Linear Systems with Sub-Gaussian Disturbances using Causal Predictors
• Data-Driven Predictive Control Towards Multi-Agent Motion Planning With Non-Parametric Closed-Loop Behavior Learning
• A new perspective on Willems' fundamental lemma: Universality of persistently exciting inputs
• A Linear Parameter-Varying Approach to Data Predictive Control
• Autonomous path following using data-driven predictive control