A predictive safety filter for learning-based control of constrained nonlinear dynamical systems
From MaRDI portal
Publication:2665095
Abstract: The transfer of reinforcement learning (RL) techniques into real-world applications is challenged by safety requirements in the presence of physical limitations. Most RL methods, in particular the most popular algorithms, do not support explicit consideration of state and input constraints. In this paper, we address this problem for nonlinear systems with continuous state and input spaces by introducing a predictive safety filter, which is able to turn a constrained dynamical system into an unconstrained safe system and to which any RL algorithm can be applied `out-of-the-box'. The predictive safety filter receives the proposed control input and decides, based on the current system state, if it can be safely applied to the real system, or if it has to be modified otherwise. Safety is thereby established by a continuously updated safety policy, which is based on a model predictive control formulation using a data-driven system model and considering state and input dependent uncertainties.
Recommendations
- Reinforcement learning control of constrained dynamic systems with uniformly ultimate boundedness stability guarantee
- Safe exploration in model-based reinforcement learning using control barrier functions
- Safe reinforcement learning for continuous spaces through Lyapunov-constrained behavior
- Safe control of nonlinear systems in LPV framework using model-based reinforcement learning
- A comprehensive survey on safe reinforcement learning
Cites work
- A comprehensive survey on safe reinforcement learning
- A General Safety Framework for Learning-Based Control in Uncertain Robotic Systems
- A quasi-infinite horizon nonlinear model predictive control scheme with guaranteed stability
- Control Barrier Function Based Quadratic Programs for Safety Critical Systems
- End-to-end training of deep visuomotor policies
- Gaussian processes for machine learning.
- Hybrid Systems: Computation and Control
- Model predictive control: recent developments and future promise
- Nonlinear Reference Tracking: An Economic Model Predictive Control Perspective
- On the implementation of an interior-point filter line-search algorithm for large-scale nonlinear programming
- Statistical learning theory: models, concepts, and results
- Tolerance regions for a multivariate normal population
Cited in
(20)- Computationally efficient infinite-horizon indefinite model predictive control with disturbance preview information
- Planning for potential: efficient safe reinforcement learning
- Reinforcement learning control of constrained dynamic systems with uniformly ultimate boundedness stability guarantee
- Blind spot detection for safe sim-to-real transfer
- A General Safety Framework for Learning-Based Control in Uncertain Robotic Systems
- A comprehensive survey on safe reinforcement learning
- Safety-aware apprenticeship learning
- Adaptive dynamic programming for optimal control of discrete‐time nonlinear system with state constraints based on control barrier function
- Safe learning-based model predictive control using the compatible models approach
- \textsc{GoSafeOpt}: scalable safe exploration for global optimization of dynamical systems
- Approximate dynamic programming for constrained linear systems: a piecewise quadratic approximation approach
- Kernel-based regularized iterative learning control of repetitive linear time-varying systems
- Stable predictive control of continuous stirred-tank reactors using deep learning
- Safety reinforcement learning control via transfer learning
- Sampled-data funnel control and its use for safe continual learning
- Safe robust multi-agent reinforcement learning with neural control barrier functions and safety attention mechanism
- A data-driven safety preserving control architecture for constrained cyber-physical systems
- Safe continual learning in model predictive control with prescribed bounds on the tracking error
- Safely learning dynamical systems
- Model predictive control: past, present, and future
This page was built for publication: A predictive safety filter for learning-based control of constrained nonlinear dynamical systems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2665095)
