Safe Reinforcement Learning Using Robust MPC
From MaRDI portal
Abstract: Reinforcement Learning (RL) has recently impressed the world with stunning results in various applications. While the potential of RL is now well-established, many critical aspects still need to be tackled, including safety and stability issues. These issues, while partially neglected by the RL community, are central to the control community which has been widely investigating them. Model Predictive Control (MPC) is one of the most successful control techniques because, among others, of its ability to provide such guarantees even for uncertain constrained systems. Since MPC is an optimization-based technique, optimality has also often been claimed. Unfortunately, the performance of MPC is highly dependent on the accuracy of the model used for predictions. In this paper, we propose to combine RL and MPC in order to exploit the advantages of both and, therefore, obtain a controller which is optimal and safe. We illustrate the results with a numerical example in simulations.
Cited in
(34)- Scalable tube model predictive control of uncertain linear systems using ellipsoidal sets
- Lagrangian-based online safe reinforcement learning for state-constrained systems
- Robust MPC with event-triggered learning for unknown linear time-varying systems
- Fast nonlinear model predictive control combining online trajectory optimization and value function regression
- Model predictive control for space robot manipulator modeled by switched systems: a data-driven approach
- Learning for MPC with stability \& safety guarantees
- Ambiguity tube MPC
- Enhancing safety in model-based reinforcement learning with high-order control barrier functions
- Safe Exploration of State and Action Spaces in Reinforcement Learning
- Data-driven safe gain-scheduling control
- The implicit rigid tube model predictive control
- Model independent dynamic predictive controller design using differential extreme learning machine for composition control in binary distillation column
- Safety-critical optimal control of discrete-time non-linear systems via policy iteration-based Q-learning
- Probabilistic reachable sets of stochastic nonlinear systems with contextual uncertainties
- Safe navigation in adversarial environments
- Multi-agent reinforcement learning via distributed MPC as a function approximator
- Learning-based state estimation and control using MHE and MPC schemes with imperfect models
- Data-driven model predictive control with reinforcement learning for linear time-invariant systems
- Sample-based robust data-enabled predictive control for safe motion planning of unknown systems
- Safe reward‐based deep reinforcement learning control for an electro‐hydraulic servo system
- Stability-constrained Markov decision processes using MPC
- Learning‐based model predictive control under value iteration with finite approximation errors
- Robust tube-based reinforcement learning control for systems with parametric uncertainty
- Closed-loop performance optimization of model predictive control with robustness guarantees
- Robust learning-based iterative model predictive control for unknown non-linear systems
- Safe learning-based model predictive control using the compatible models approach
- Heterogeneous optimal formation control of nonlinear multi-agent systems with unknown dynamics by safe reinforcement learning
- Safe fixed-time reinforcement learning for nonlinear zero-sum games with obstacle avoidance awareness
- Intrinsic separation principles
- Data-driven ensemble optimal compensation control for partially known delayed and persistently disturbed nonlinear systems
- Safe reinforcement learning: A control barrier function optimization approach
- Safe control of nonlinear systems in LPV framework using model-based reinforcement learning
- Explicit explore, exploit, or escape \((E^4)\): near-optimal safety-constrained reinforcement learning in polynomial time
- Safety-constrained reinforcement learning with a distributional safety critic
This page was built for publication: Safe Reinforcement Learning Using Robust MPC
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4957610)
