Online learning‐based model predictive control with Gaussian process models and stability guarantees
From MaRDI portal
Abstract: Model predictive control allows to provide high performance and safety guarantees in the form of constraint satisfaction. These properties, however, can be satisfied only if the underlying model, used for prediction, of the controlled process is sufficiently accurate. One way to address this challenge is by data-driven and machine learning approaches, such as Gaussian processes, that allow to refine the model online during operation. We present a combination of an output feedback model predictive control scheme and a Gaussian process-based prediction model that is capable of efficient online learning. To this end, the concept of evolving Gaussian processes is combined with recursive posterior prediction updates. The presented approach guarantees recursive constraint satisfaction and input-to-state stability with respect to the model-plant mismatch. Simulation studies underline that the Gaussian process prediction model can be successfully and efficiently learned online. The resulting computational load is significantly reduced via the combination of the recursive update procedure and by limiting the number of training data points while maintaining good performance.
Recommendations
- Online learning stochastic model predictive control of linear uncertain systems
- Online learning constrained model predictive control based on double prediction
- Nonlinear learning‐based model predictive control supporting state and input dependent model uncertainty estimates
- High-probability stable Gaussian process-supported model predictive control for Lur'e systems
Cites work
- A General Safety Framework for Learning-Based Control in Uncertain Robotic Systems
- Constrained model predictive control: Stability and optimality
- Dynamic systems identification with Gaussian processes
- Fault tolerant control using Gaussian processes and model predictive control
- Gaussian processes for machine learning.
- Gaussian processes for time-series modelling
- Handbook of model predictive control
- Information-Theoretic Regret Bounds for Gaussian Process Optimization in the Bandit Setting
- Input-to-state stability: a unifying framework for robust model predictive control
- Min-max feedback model predictive control for constrained linear systems
- Model predictive control: recent developments and future promise
- Modelling and control of dynamic systems using Gaussian process models
- On the stability of constrained MPC without terminal constraint
- Provably safe and robust learning-based model predictive control
- Robust output feedback model predictive control of constrained linear systems
- Sparse spectrum Gaussian process regression
- Switching and Learning in Feedback Systems
Cited in
(18)- Online model learning of buildings using stochastic hybrid systems based on Gaussian processes
- High-probability stable Gaussian process-supported model predictive control for Lur'e systems
- Online learning constrained model predictive control based on double prediction
- Nonlinear learning‐based model predictive control supporting state and input dependent model uncertainty estimates
- Physics-informed online learning of gray-box models by moving horizon estimation
- Zero-order optimization for Gaussian process-based model predictive control
- Online learning stochastic model predictive control of linear uncertain systems
- Learning‐based adaptive‐scenario‐tree model predictive control with improved probabilistic safety using robust Bayesian neural networks
- Online learning-based event-triggered model predictive control with shrinking prediction horizon for perturbed nonlinear systems
- Efficient implementation of Gaussian process-based predictive control by quadratic programming
- Robust learning-based iterative model predictive control for unknown non-linear systems
- Safe continual learning in model predictive control with prescribed bounds on the tracking error
- Learning for predictive control: a dual Gaussian process approach
- Flexible development and evaluation of machine-learning-supported optimal control and estimation methods via HILO-MPC
- Model predictive control: past, present, and future
- A GP-enhanced MPC framework with information-driven adaptation and risk-awareness for UAV trajectory tracking
- Data/moment-driven approaches for fast predictive control of collective dynamics
- A robust and adaptive MPC formulation for Gaussian process models
This page was built for publication: Online learning‐based model predictive control with Gaussian process models and stability guarantees
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6060775)
