Pith. sign in

REVIEW 4 major objections 3 minor 1 cited by

Local Bayesian Optimization for Controller Tuning with Crash Constraints

T0 review · 4 major / 3 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A local Bayesian optimizer that adds virtual points at crash locations tunes 2- to 8-parameter controllers in a few hundred evaluations, and improves hardware PI tracking by roughly 33% and cascaded PI by 50%.

desk verdict Useful, incremental controller-tuning paper with real hardware evidence; the crash-penalty heuristic is the main soft spot but not a reason to reject. read the letter →

arxiv 2411.16267 v1 pith:TEV4FYW7 submitted 2024-11-25 eess.SY cs.LGcs.SY

Alexander von Rohr , David Stenger , Dominik Scheurenberg , Sebastian Trimpe This is my paper · ORCID
classification eess.SYcs.LGcs.SY
keywords BayesianoptimizationcontrollertuningcrashconstraintslocalGaussianprocessvirtualdatapointsgradientinformationcoupledtanksystem
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Controller tuning often fails outright for some parameter settings: the closed-loop system crashes and no performance value is returned. This paper proposes VDP-GIBO, a local Bayesian optimization method that handles these crash constraints by inserting a virtual data point with an adaptive penalty at each crash location into the Gaussian process model. The virtual points steer the learned gradient away from infeasible regions, so the standard GIBO gradient-update loop can keep improving the controller. In a simulated coupled tank system the method tunes 2-, 5-, 6- and 8-parameter PI, cascaded PI, LQI and MPC controllers within tens to a few hundred evaluations, and on real hardware it improves PI tracking by about 33% and cascaded PI by about 50%.

What carries the argument

The central object is the augmented Gaussian process posterior built from both real and virtual observations. When an evaluation at x̂_i crashes, the algorithm adds the point (x̂_i, ŷ_i) with ŷ_i = max(μ_D(x̂_i), μ_D(x*)) + β√k_D(x̂_i) and β=3, which lies above the posterior mean at the current iterate and at the crash site; this creates a ridge that tilts the GP gradient away from the infeasible region. The gradient update x* ← x* − η_k ∇μ_hatD(x_k) uses the posterior mean of the derivative, and the step size η_k is normalized by the GP lengthscale via Eq. (5). The next batch of evaluations minimizes the total trace variance of the gradient posterior (Eq. (2)–(3)), concentrating queries where the gradient estimate is most uncertain, and a reset rule returns the iterate to the best feasible point seen so far if an update lands in the crash region.

What would settle it

On a one-dimensional synthetic problem with feasible set [0, 0.5] and an objective that decreases steeply toward a minimum just inside the boundary at 0.5, run VDP-GIBO with β=3 from any feasible start. If the algorithm ever proposes a second evaluation at a crash location (x > 0.5) without the virtual point's gradient pointing outward — or if the reset rule triggers at the same iterate more than once — the heuristic is insufficient in that regime. More directly, compute the angle between ∇μ_hatD(x̂_i) and the outward normal of the feasible set after inserting the virtual point; the method requires that angle to exceed 90° at every crash location.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that crash constraints can be handled at the GP modeling level rather than at the acquisition-function level: virtual observations with value ŷ_i = max(μ_D(x̂_i), μ_D(x*)) + β√k_D(x̂_i) modify the posterior so that the GP gradient at the current iterate points away from observed crash locations, allowing the GIBO update rule to keep making descent steps. The authors demonstrate this on five simulated test cases with deterministic crashes, showing that VDP-GIBO matches or beats random search on the easy 2D problems and clearly outperforms it on the 5-, 6-, and 8-dimensional problems within the same evaluation budget. They further validate the approach on a physical coupled tank, where tuning runs complete in 33 to 54 evaluations and improve closed-loop tracking error by roughly one-third (PI) to one-half (cascaded PI).

Load-bearing premise

The whole crash-handling scheme rests on the heuristic that a virtual point at the crash location with value max(μ_D(x̂_i), μ_D(x*)) + 3√k_D(x̂_i) is large enough to make the learned gradient point away from the infeasible region; if the real objective's slope toward the crash region is steeper than this penalty, the algorithm can keep proposing infeasible points and stall.

Editorial extensions

If this is right

  • Because crash handling is confined to the GP update, the virtual-data mechanism can in principle be dropped into other BO variants that use a Gaussian process, without redesigning the acquisition function.
  • Controller tuning with many parameters becomes feasible: the paper demonstrates 8-parameter LQI tuning in about 200 evaluations, a regime where random search makes little progress.
  • Since the search is local and the penalty discourages revisiting crashes, tuning sessions on physical systems will rarely trigger emergency stops; in the hardware experiments almost all evaluations stayed feasible.
  • The fixed hyperparameter set (including β=3, the kernel lengthscale and the step-size schedule) worked across all test cases, indicating the method is not strongly sensitive to these choices.
  • As long as crashes are deterministic, the reset rule guarantees a feasible parameterization after every update, so the tuning process never ends in an infeasible state.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The virtual-penalty heuristic could be stress-tested on synthetic functions with known steep gradients toward the feasible boundary; one would predict that β=3 sometimes fails to turn the GP gradient outward, which would indicate when a larger β or a different penalty shape is needed.
  • The paper assumes deterministic crashes; if real systems can crash stochastically, the reset rule no longer guarantees a feasible iterate, and the algorithm would likely need a probabilistic feasibility model to remain safe.
  • The same virtual-data strategy could be combined with global acquisition functions like expected improvement to handle crashes in non-local search; the paper only demonstrates it inside the local GIBO loop, so the transfer remains untested.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 3 minor

Summary. The paper proposes VDP-GIBO, a local Bayesian optimization method for controller tuning under crash constraints, where failed evaluations do not return a function value. The method extends the authors' earlier GIBO algorithm by adding virtual data points at crash locations with an adaptive penalty (Eq. 4), by evaluating batches that minimize the total variance of the GP posterior gradient, and by resetting to a known feasible point when an update becomes infeasible (Algorithm 1). The method is evaluated on a simulated coupled-tank system with PI, cascaded PI, MPC+EKF, and LQI controllers (2, 5, 6, and 8 tuning parameters, respectively) against a random-search baseline, and on a hardware testbed for PI and cascaded PI control. The paper reports that VDP-GIBO solves the higher-dimensional problems within tens to a few hundred evaluations and improves hardware PI tracking by about 33% and cascaded PI by about 50%.

Significance. If the reported results hold, VDP-GIBO would be a practically useful tool for controller tuning in settings where global Bayesian optimization is too sample-hungry and where crashed experiments are common. The paper has several strengths: simulation results are averaged over ten random seeds with randomized initializations; the same hyperparameters are used across all problems; and the hardware experiments demonstrate the crash-constraint machinery on a real system. The proposed integration of virtual data points into a local BO framework is clean and easy to combine with different acquisition procedures. However, the novelty is incremental relative to prior work by the same group (GIBO and VDP-BO), and the load-bearing crash-handling heuristic in Eq. (4) is not supported by a derivation or sensitivity analysis. The empirical evaluation compares only against random search, and the hardware results are single-run point estimates. These gaps limit the strength of the data-efficiency claim in its current form.

major comments (4)
  1. [4.1, Eq. (4)] The virtual penalty in Eq. (4) is the only mechanism that converts crashed evaluations into gradient information, yet the paper does not establish that the posterior gradient ∇μ_𝒟hat(x*) reliably points away from the crash region. The lower bound max(μ_D(x*), μ_D(x̂_i)) only ensures a non-negative secant slope between x* and each crash location; it does not control the direction of the GP gradient, and re-optimizing hyperparameters in Algorithm 1 (line 14) can alter the posterior after the virtual points are inserted. Please add a sensitivity analysis over β (e.g., 1, 3, 10), an ablation replacing Eq. (4) with a fixed penalty, and a diagnostic showing the direction of the gradient update after a crash. Without such evidence, the data-efficiency claim for crash-constrained problems rests on a single illustrative example (Fig. 2) and on the specific β=3 value used in the experiments.
  2. [6, Fig. 7] The hardware results are based on one run per configuration. The reported 33% and 50% improvements are point estimates from single trajectories; Fig. 7 shows no error bars or repeated experiments. Because these results are the main evidence for the practical value of the method, please report multiple runs with variability, or clearly label the results as single-session illustrations and adjust the strength of the conclusions accordingly.
  3. [5.3, Tab. 2] The paper states that the hyperparameters (β=3, L=0.25I, σ_f=0.5, μ=1) were chosen manually from initial experimentation with the PI controller and then concludes that VDP-GIBO is "not very sensitive" to the choice of hyperparameters. This conclusion is not supported by any sensitivity study. Please provide such a study (e.g., varying β and L over a range and reporting performance on at least one problem) or replace the claim with the more limited statement that the same hyperparameters produced good results on the tested problems.
  4. [5.3, Fig. 4] The only baseline is random search. While this supports the specific claim that VDP-GIBO outperforms random search on the higher-dimensional problems, it does not support the broader data-efficiency claim relative to established Bayesian optimization variants with crash handling. Please add comparisons to at least one alternative, such as global BO with a fixed penalty (as in [5]) or a constrained BO method (as in [7]), or clearly scope the contribution as "better than random search" rather than "data-efficient" in a general sense.
minor comments (3)
  1. [Algorithm 1, line 9] The gradient update writes "∇μ_𝒟(x_k)" but the loop variable is x*; please use consistent notation (e.g., ∇μ_𝒟hat(x*)).
  2. [Fig. 4 caption] The caption says "4-(cascaded PI)" but Tab. 1 and Section 5.2 identify the cascaded PI case as having 5 parameters; please correct the inconsistency.
  3. [Section 6, first experiment] The text says "eight iterations, corresponding to 33 evaluations" for a 2-D problem with batch size b=d+1=3; a reader would expect 8 iterations to involve at least 25 evaluations. Please clarify how the total of 33 is counted (e.g., including the initial evaluation and the per-update evaluation).

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: VDP-GIBO reuses the authors' earlier GIBO and virtual-data-point components, but the paper's data-efficiency claims are validated on new simulations and hardware experiments against an external random-search baseline.

full rationale

The derivation chain does not equate an output with an input. Equation (2) and Algorithm 1's design-of-experiments step are taken from Müller et al. [8], and the virtual penalty (4) is explicitly attributed to Stenger and Abel [11] ('Following Stenger and Abel[11]...'); the paper never claims to derive (4) from first principles. The lower-bound argument around (4) only motivates the heuristic; it is illustrated in Fig. 2 and empirically set with beta=3, and the paper acknowledges that determining beta can be difficult. The central claim—data-efficient tuning under crash constraints—is supported by fresh evidence: five simulation settings compared with a uniform random-search baseline (Fig. 4) and three hardware tuning runs with ca. 33% and ca. 50% measured improvements (Fig. 7). Those outcomes are not forced by fitting the GP to the same targets, and all GP hyperparameters are fixed across experiments. The self-citations to [8] and [11] are cumulative method reuse, not load-bearing appeals to an unverified theorem; the 'unique minimum' remark citing [11] is peripheral. The under-analyzed virtual-penalty mechanism is a robustness/correctness limitation, not a circular step.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

No new physical entities are postulated; virtual data points are an algorithmic construct, not an invented entity in the sense of a new force or dimension. The free parameters are the algorithm's hyperparameters, manually selected, and the axioms are the standard BO modeling assumptions plus the crash-constraint heuristic.

free parameters (6)
  • β (virtual penalty scaling) = 3
    Chosen empirically in Section 4.1; controls how much worse than the GP prediction a crash is assumed to be.
  • GP prior mean μ(x) = 1
    Constant prior mean used in all experiments (Tab. 2).
  • GP signal variance σ_f = 0.5
    Fixed kernel amplitude in Tab. 2, chosen from initial PI experiments.
  • GP lengthscale matrix L = 0.25 I
    Fixed isotropic lengthscale in Tab. 2; controls smoothness assumptions.
  • Step size η_k = 0.25 down to 0.125 with cosine decay
    Manual schedule in Tab. 2; affects convergence speed.
  • Batch size b_k = d + 1
    Number of design-of-experiment evaluations per iteration; d is the search dimension.
assumptions (5)
  • domain assumption The performance function f is a sample from a GP with at least once differentiable mean and twice differentiable covariance (Assumption 1).
    Standard in BO but not verifiable for the real plant; enables analytic posterior gradients.
  • ad hoc to paper The virtual data point penalty of Eq. (4) with β=3 shapes the GP posterior so the estimated gradient points away from crashes.
    Heuristic from [11]; no proof that the modified GP has the correct descent direction near the crash boundary.
  • domain assumption Controller tuning problems frequently have a unique minimum (Section 1, citing [11]).
    Justifies local search; if false, VDP-GIBO may converge to an inferior local optimum.
  • domain assumption Crashes are deterministic, so resetting to a known feasible evaluation guarantees a viable parameterization (Section 4.2).
    Stated explicitly; if crashes are stochastic, the reset may not be feasible.
  • domain assumption A Gaussian kernel with fixed lengthscales L=0.25I and σ_f=0.5 adequately models the objective (Section 4.2, Tab. 2).
    Kernel and hyperparameters are fixed across all problems without per-problem fitting; assumes similar smoothness across systems.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Local Bayesian Optimization for Controller Tuning with Crash Constraints." pith.science (2026). https://pith.science/paper/TEV4FYW7

@misc{pith2026241116267,
  author       = {Pith},
  title        = {Pith review of: Local Bayesian Optimization for Controller Tuning with Crash Constraints},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TEV4FYW7}},
  note         = {Machine review of arXiv:2411.16267}
}
read the original abstract

Controller tuning is crucial for closed-loop performance but often involves manual adjustments. Although Bayesian optimization (BO) has been established as a data-efficient method for automated tuning, applying it to large and high-dimensional search spaces remains challenging. We extend a recently proposed local variant of BO to include crash constraints, where the controller can only be successfully evaluated in an a-priori unknown feasible region. We demonstrate the efficiency of the proposed method through simulations and hardware experiments. Our findings showcase the potential of local BO to enhance controller performance and reduce the time and resources necessary for tuning.

Figures

Figures reproduced from arXiv: 2411.16267 by the authors.

Figure 1
Figure 1. The controller tuning process with BO. The objective 𝑓 is evaluated in closed-loop. The controller 𝜋𝑥 has tuning parameters 𝑥 ∈ 𝒳 and BO searches for the optimal parameterization. No function value is available if an experiment crashes, 𝑥 ̸∈ 𝒳𝑆 ⊆ 𝒳. prior knowledge about the plant and data collected during its operation. An emerging approach in controller tuning is Bayesian Optimization (BO), which is particularly w… view at source ↗
Figure 2
Figure 2. Left: A Gaussian process posterior (top) and its derivative (bottom). Right: The posterior with an additional virtual observation in the crash region 𝒳𝐶 . The crashed evaluation (red cross) cannot be evaluated, and a virtual observation is added instead. In this example, the virtual data point modifies the posterior such that the minimum of the posterior is not inside the infeasible region 𝒳𝐶 , and the gradient poin… view at source ↗
Figure 3
Figure 3. Diagram of the coupled tank system, with controllable pump and valves. controller tuning tasks. We also introduce the control algo￾rithms under consideration and present the results of the simulated controller tuning process using VDP-GIBO. 5.1 Process Description The subject of examination in this study is a coupled tank system, illustrated in [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Simulation results on crash constrained controller tuning problems: [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Evaluations in the parameter space for PI (𝑉2 < 8 l) (left) and PI (𝑉2 < 7 l) (right). We show the first eight improvement steps and the corresponding evaluation locations 𝑋, improvement steps 𝑋* and crashes 𝑋^. Due to the tighter constraints on the right, the feasible…
Figure 6
Figure 6. Figure 6: Analogous to the description of the simulation model, water is pumped into tank B2 by pump P1. The volumetric flow generated by the pump as a function of the control input is measured using a flow sensor. The pneumatic valve V1 controls the volume flow from B2 to B3. S…
Figure 7
Figure 7. Figure 7: Experimental results for crash-constrained PI control: [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 8
Figure 8. Figure 8: Time domain of the PI controller: Step response of 𝑉4 (purple) and 𝑉2 (petrol) with the PI controller. After tuning, the state tracks the desired reference (black) significantly better. Darker colors indicate the behavior later in the tuning process. The state 𝑉2 stays…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Copy First, Translate Later: Interpreting Translation Dynamics in Multilingual Pretraining

    cs.CL 2026-04 unverdicted novelty 7.0 of 10

    SLO-Guard, a crash-aware two-phase autotuner for vLLM serving, achieves no best-latency improvement over random search but demonstrates more consistent budget allocation across 150 trials on Qwen2-1.5B/A100.

Reference graph

Works this paper leans on

33 extracted references · 31 canonical work pages · cited by 1 Pith paper

  1. [5]

    Automatic LQR tuning based on Gaussian process global optimization

    Alonso Marco, Philipp Hennig, Jeannette Bohg, Stefan Schaal, and Sebastian Trimpe. Automatic LQR tuning based on Gaussian process global optimization. In2016 IEEE International Conference on Robotics and Automation (ICRA), pages 270–277, 2016

    work page 2016

  2. [7]

    Robot learn- ing with crash constraints.IEEE Robotics and Automation Letters, 6(2):1439–1446, 2021

    Alonso Marco, Dominik Baumann, Majid Khadiv, Philipp Hennig, Ludovic Righetti, and Sebastian Trimpe. Robot learn- ing with crash constraints.IEEE Robotics and Automation Letters, 6(2):1439–1446, 2021

    work page 2021

  3. [1]

    An overview of control performance assessment technology and industrial applications.Control Engineering Practice, 14(5):441–466, 2006

    Mohieddine Jelali. An overview of control performance assessment technology and industrial applications.Control Engineering Practice, 14(5):441–466, 2006

    work page 2006

  4. [2]

    A survey on policy search algorithms for learning robot controllers in a handful of trials.IEEE Transactions on Robotics, 36(2): 328–347, April 2020

    Konstantinos Chatzilygeroudis, Vassilis Vassiliades, Freek Stulp, Sylvain Calinon, and Jean-Baptiste Mouret. A survey on policy search algorithms for learning robot controllers in a handful of trials.IEEE Transactions on Robotics, 36(2): 328–347, April 2020

    work page 2020

  5. [3]

    Paulson, Farshud Sorourifar, and Ali Mesbah

    Joel A. Paulson, Farshud Sorourifar, and Ali Mesbah. A tutorial on derivative-free policy learning methods for inter- pretable controller representations. In2023 American Control Conference (ACC), pages 1295–1306, 2023

    work page 2023

  6. [4]

    Gaus- sian process optimization with failures: classification and convergence proof

    François Bachoc, Céline Helbert, and Victor Picheny. Gaus- sian process optimization with failures: classification and convergence proof. Journal of Global Optimization, 78(3): 483–506, jul 2020

    work page 2020

  7. [6]

    Bayesian optimization for learning gaits under uncertainty

    Roberto Calandra, André Seyfarth, Jan Peters, and Marc Pe- ter Deisenroth. Bayesian optimization for learning gaits under uncertainty. Annals of Mathematics and Artificial Intelligence, 76(1-2):5–23, 2016

    work page 2016

  8. [8]

    Local policy search with Bayesian optimization

    Sarah Müller, Alexander von Rohr, and Sebastian Trimpe. Local policy search with Bayesian optimization. InAdvances in Neural Information Processing Systems, volume 34, pages 20708–20720, 2021

    work page 2021

Show all 33 references
  1. [9]

    Local Bayesian optimization via maximizing probability of descent

    Quan Nguyen, Kaiwen Wu, Jacob Gardner, and Roman Gar- nett. Local Bayesian optimization via maximizing probability of descent. InAdvances in Neural Information Processing Systems, volume 35, pages 13190–13202, 2022

    2022

  2. [10]

    Kaiwen Wu, Kyurae Kim, Roman Garnett, and Jacob R. Gardner. The behavior and convergence of local bayesian optimization. In Thirty-seventh Conference on Neural Infor- mation Processing Systems, 2023

    2023

  3. [11]

    Benchmark of bayesian optimization and metaheuristics for control engineering A

    David Stenger and Dirk Abel. Benchmark of bayesian optimization and metaheuristics for control engineering A. von Rohr et al., Local Bayesian Optimization for Controller Tuning 11 tuning problems with crash constraints.arXiv preprint arXiv:2211.02571, 2022

    2022 arXiv

  4. [12]

    Gaussian process optimization in the bandit setting: No regret and experimental design

    Niranjan Srinivas, Andreas Krause, Sham Kakade, and Matthias Seeger. Gaussian process optimization in the bandit setting: No regret and experimental design. InInternational Conference on Machine Learning, pages 1015–1022, 2010

    2010

  5. [13]

    Model-predictive control with stochastic collision avoidance using Bayesian policy optimization

    Olov Andersson, Mariusz Wzorek, Piotr Rudol, and Patrick Doherty. Model-predictive control with stochastic collision avoidance using Bayesian policy optimization. In2016 IEEE International Conference on Robotics and Automation (ICRA), pages 4597–4604, 2016

    2016

  6. [14]

    Controller parameter optimization for com- plex industrial system with uncertainties.Measurement and Control, 52(7-8):888–895, 2019

    Heping Chen, Seth Bowels, Biao Zhang, and Thomas Fuhlbrigge. Controller parameter optimization for com- plex industrial system with uncertainties.Measurement and Control, 52(7-8):888–895, 2019

    2019

  7. [15]

    Weak in the NEES?: Auto-tuning Kalman filters with Bayesian optimization

    Zhaozhong Chen, Christoffer Heckman, Simon Julier, and Nisar Ahmed. Weak in the NEES?: Auto-tuning Kalman filters with Bayesian optimization. In2018 21st International Conference on Information Fusion (FUSION), pages 1072– 1079, 2018

    2018

  8. [16]

    Min- max hyperparameter tuning, with application to fault detec- tion

    Julien Marzat, Hélène Piet-Lahanier, and Eric Walter. Min- max hyperparameter tuning, with application to fault detec- tion. IFAC Proceedings Volumes, 44(1):12904–12909, 2011. 18th IFAC World Congress

    2011

  9. [17]

    Smith, Alisa Rupenyan, and John Lygeros

    Mohammad Khosravi, Varsha Behrunani, Roy S. Smith, Alisa Rupenyan, and John Lygeros. Cascade control: Data- driven tuning approach based on Bayesian optimization. IFAC-PapersOnLine, 53(2):382–387, 2020. 21st IFAC World Congress

    2020

  10. [18]

    Joint constrained Bayesian optimization of planning, guidance, control, and state estimation of an autonomous underwater vehicle

    David Stenger, Maximilian Nitsch, and Dirk Abel. Joint constrained Bayesian optimization of planning, guidance, control, and state estimation of an autonomous underwater vehicle. In 2022 European Control Conference (ECC), pages 1982–1987, 2022

    2022

  11. [19]

    Safe contextual Bayesian optimization for sustainable room temperature PID control tuning

    Marcello Fiducioso, Sebastian Curi, Benedikt Schumacher, Markus Gwerder, and Andreas Krause. Safe contextual Bayesian optimization for sustainable room temperature PID control tuning. InProceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence,...

    2019

  12. [20]

    Smith, John Lygeros, and Alisa Rupenyan

    Mohammad Khosravi, Christopher König, Markus Maier, Roy S. Smith, John Lygeros, and Alisa Rupenyan. Safety- aware cascade controller tuning using constrained Bayesian optimization. IEEE Transactions on Industrial Electronics, 70 (2):2128–2138, 2023

    2023

  13. [21]

    Schoellig

    Felix Berkenkamp, Andreas Krause, and Angela P. Schoellig. Bayesian optimization with safety constraints: safe and automatic parameter tuning in robotics.Machine Learning, 2021

    2021

  14. [22]

    What is the best way to optimally parameterize the MPC cost function for vehicle guidance? Mathematics, 11(2), 2023

    David Stenger, Robert Ritschel, Felix Krabbes, Rick Voßwinkel, and Hendrik Richter. What is the best way to optimally parameterize the MPC cost function for vehicle guidance? Mathematics, 11(2), 2023

    2023

  15. [23]

    Bonzanini, Victor Miller, and Ali Mesbah

    Georgios Makrygiorgos, Angelo D. Bonzanini, Victor Miller, and Ali Mesbah. Performance-oriented model learning for control via multi-objective Bayesian optimization.Computers & Chemical Engineering, 162:107770, 2022

    2022

  16. [24]

    Noisy-input entropy search for efficient robust Bayesian optimization

    Lukas Fröhlich, Edgar Klenske, Julia Vinogradska, Christian Daniel, and Melanie Zeilinger. Noisy-input entropy search for efficient robust Bayesian optimization. In Silvia Chiappa and Roberto Calandra, editors,Proceedings of the Twenty Third International Conference on Artific...

    2020

  17. [25]

    Paulson, Georgios Makrygiorgos, and Ali Mesbah

    Joel A. Paulson, Georgios Makrygiorgos, and Ali Mesbah. Adversarially robust Bayesian optimization for efficient auto- tuning of generic control structures under uncertainty.AIChE Journal, 68(6):e17591, 2022

    2022

  18. [26]

    Mengjia Zhu, Dario Piga, and Alberto Bemporad. C-GLISp: Preference-based global optimization under unknown con- straints with applications to controller calibration.IEEE Transactions on Control Systems Technology, pages 1–12, 2021

    2021

  19. [27]

    On controller tuning with time-varying Bayesian optimiza- tion

    Paul Brunzema, Alexander Von Rohr, and Sebastian Trimpe. On controller tuning with time-varying Bayesian optimiza- tion. In 2022 IEEE 61st Conference on Decision and Control (CDC), pages 4046–4052, 2022

    2022

  20. [28]

    Robust parametrization of a model predictive controller for a CNC machining center using Bayesian optimization.IFAC- PapersOnLine, 53(2):10388–10394, 2020

    David Stenger, Muzaffer Ay, and Dirk Abel. Robust parametrization of a model predictive controller for a CNC machining center using Bayesian optimization.IFAC- PapersOnLine, 53(2):10388–10394, 2020. 21st IFAC World Congress

    2020

  21. [29]

    Cambridge University Press, 2023

    Roman Garnett.Bayesian Optimization. Cambridge University Press, 2023

    2023

  22. [30]

    Rasmussen and CKI

    CE. Rasmussen and CKI. Williams.Gaussian Processes for Machine Learning. MIT Press, 2006

    2006

  23. [31]

    Evaluation of data enhanced model predictive control for a coupled tank system

    Dominik Scheurenberg, Sebastian Stemmler, and Dirk Abel. Evaluation of data enhanced model predictive control for a coupled tank system. In2023 IEEE Conference on Control Technology and Applications (CCTA), pages 79–84, 2023

    2023

  24. [32]

    GoSafe: Globally optimal safe robot learning

    Dominik Baumann, Alonso Marco, Matteo Turchetta, and Sebastian Trimpe. GoSafe: Globally optimal safe robot learning. In2021 IEEE International Conference on Robotics and Automation (ICRA), pages 4452–4458, 2021

    2021

  25. [33]

    Formal safety net control using back- ward reachability analysis.IEEE Transactions on Automatic Control, 67(11):5698–5713, 2022

    Bastian Schürmann, Moritz Klischat, Niklas Kochdumper, and Matthias Althoff. Formal safety net control using back- ward reachability analysis.IEEE Transactions on Automatic Control, 67(11):5698–5713, 2022

    2022

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.