• Corpus ID: 238856830

VABO: Violation-Aware Bayesian Optimization for Closed-Loop Control Performance Optimization with Unmodeled Constraints

@article{Xu2021VABOVB,
  title={VABO: Violation-Aware Bayesian Optimization for Closed-Loop Control Performance Optimization with Unmodeled Constraints},
  author={Wenjie Xu and Colin Neil Jones and Bratislav Svetozarevic and Christopher R. Laughman and Ankush Chakrabarty},
  journal={ArXiv},
  year={2021},
  volume={abs/2110.07479}
}
We study the problem of performance optimization of closed-loop control systems with unmodeled dynamics. Bayesian optimization (BO) has been demonstrated effective for improving closed-loop performance by automatically tuning controller gains or reference setpoints in a model-free manner. However, BO methods have rarely been tested on dynamical systems with unmodeled constraints. In this paper, we propose a violation-aware BO algorithm (VABO) that optimizes closed-loop performance while… 
View PDF on arXiv
View With Semantic Reader

Figures from this paper

References

SHOWING 1-10 OF 30 REFERENCES
Data-Driven Scenario Optimization for Automated Controller Tuning With Probabilistic Performance Guarantees
TLDR
This letter uses recent advances in non-convex scenario theory to provide a distribution-free bound on the probability of the closed-loop performance measures in a black-box optimization problem under uncertainty for automated controller tuning.
Performance-Oriented Model Learning for Data-Driven MPC Design
TLDR
This letter applies for the first time the above “identification for control” rationale to hierarchical MPC using data-driven methods and Bayesian optimization.
Parameter Optimization for Learning-based Control of Control-Affine Systems
TLDR
This work reformulates the parameter tuning problem using robust optimization with performance constraints based on Lyapunov theory and derives a with high probability optimal method for control parameter tuning.
Constrained Bayesian Optimization with Particle Swarms for Safe Adaptive Controller Tuning
Goal-driven dynamics learning via Bayesian optimization
TLDR
This work uses Bayesian optimization in an active learning framework where a locally linear dynamics model is learned with the intent of maximizing the control performance, and used in conjunction with optimal control schemes to efficiently design a controller for a given task.
Safe controller optimization for quadrotors with Gaussian processes
TLDR
Experimental results on a quadrotor vehicle indicate that the proposed SafeOpt algorithm enables fast, automatic, and safe optimization of controller parameters without human intervention.
Accelerating self-optimization control of refrigerant cycles with Bayesian optimization and adaptive moment estimation
Controller Tuning by Bayesian Optimization An Application to a Heat Pump
TLDR
This paper derives a physical-based model for the heat pump and identifies the unknown parameters using gray-box identification methods and tunes the controller parameters in simulation for optimal performance while considering safety constraints of the system.
Stagewise Safe Bayesian Optimization with Gaussian Processes
TLDR
An efficient safe Bayesian optimization algorithm is developed, StageOpt, that separates safe region expansion and utility function maximization into two distinct stages and provides theoretical guarantees for both the satisfaction of safety constraints as well as convergence to the optimal utility value.
Bayesian Optimization with Unknown Constraints
TLDR
This paper studies Bayesian optimization for constrained problems in the general case that noise may be present in the constraint functions, and the objective and constraints may be evaluated independently.
...
1
2
3
...