Sensitivity-Based Robust NMPC for Close-Proximity Offshore Wind Turbine Inspection with a Tilted Multirotor

Giuseppe Silano; Martin Saska

arxiv: 2605.07771 · v1 · submitted 2026-05-08 · 💻 cs.RO

Sensitivity-Based Robust NMPC for Close-Proximity Offshore Wind Turbine Inspection with a Tilted Multirotor

Giuseppe Silano , Martin Saska This is my paper

Pith reviewed 2026-05-11 02:29 UTC · model grok-4.3

classification 💻 cs.RO

keywords robust NMPCsensitivity analysismultirotor controloffshore wind turbine inspectionconstraint tighteningmodel uncertaintytilted multirotorclose-proximity flight

0 comments

The pith

Sensitivity-based robust NMPC eliminates tower clearance violations for tilted multirotor wind-turbine inspection under model mismatch.

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

The paper develops a robust version of nonlinear model predictive control that keeps a tilted multirotor safely away from the tower of an offshore wind turbine during close-proximity inspection. When the drone's mass, inertia, thrust effectiveness, drag or the surrounding wind conditions deviate from the values assumed in the model, ordinary NMPC can let the vehicle fly too close to the large cylindrical structure. The method computes first-order state sensitivities with respect to these uncertain parameters, uses them to calculate an online margin that tightens the clearance constraint, and adds a separate stage-dependent term for wind gusts. Monte-Carlo simulations with 500 different uncertainty combinations on a demanding helical flight path show that the robust controller produces no clearance violations, unlike the nominal controller, while only modestly increasing the time needed to solve each optimization problem.

Core claim

The central discovery is that augmenting a standard NMPC formulation with online constraint tightening based on first-order parametric sensitivities of the predicted state trajectory allows the controller to guarantee satisfaction of the tower-clearance constraint for all parameter values inside a bounded uncertainty set. Gust disturbances receive an additional additive margin that depends on the stage in the horizon. Because these margins are evaluated outside the optimization loop, the underlying quadratic program retains exactly the same structure, variables, and constraints as the nominal problem. Closed-loop Monte-Carlo evaluation over five hundred uncertainty realizations on a helical,

What carries the argument

online constraint tightening driven by first-order parametric state sensitivities that compute a structured uncertainty margin for the tower-clearance constraint

If this is right

The receding-horizon optimization remains a standard quadratic program of unchanged size and structure.
Clearance safety is maintained across the tested range of mass, inertia, thrust effectiveness, drag, and wind variations.
Only a moderate extra computation time is required for the sensitivity propagation and margin calculation.
The same helical inspection trajectory that causes violations under nominal control stays feasible under the robust version.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The technique could extend to other proximity-constrained aerial tasks such as power-line or bridge inspection.
Because only first-order sensitivities are used, the margins may become conservative for large uncertainty levels; second-order corrections could be investigated as an extension.
The separation of uncertainty handling from the core optimizer allows direct reuse of existing NMPC solvers and warm-starting methods.
Real-time embedded deployment on the tilted multirotor remains feasible given the reported modest solve-time overhead.

Load-bearing premise

First-order parametric state sensitivities provide a sufficient structured-uncertainty margin for the tower-clearance constraint when mass, inertia, thrust effectiveness, drag, and wind conditions differ from nominal values.

What would settle it

A Monte-Carlo trial on the boundary-critical helical inspection trajectory in which any uncertainty realization produces a tower clearance below the required minimum under the robust NMPC would falsify the claim that the sensitivity margins suffice.

Figures

Figures reproduced from arXiv: 2605.07771 by Giuseppe Silano, Martin Saska.

**Figure 2.** Figure 2: Signed clearance residuals s(t) = dmin − dT (p(t)) over 500 Monte-Carlo trials; s(t) ≤ 0 indicates constraint satisfaction. The robustified controller keeps the entire distribution strictly below the safety threshold (dashed line). ative overhead rather than embedded real-time feasibility; deployment will rely on a C-code-generated NMPC (e.g., via acados [16]) and is left to future work. VIII. CONCLUSIONS… view at source ↗

read the original abstract

Close-proximity offshore wind turbine inspection requires strict clearance control around large cylindrical structures under wind and model mismatch. Nominal Nonlinear Model Predictive Control (NMPC) may violate safety constraints when mass, inertia, thrust effectiveness, drag, or wind conditions differ from nominal assumptions. We propose a sensitivity-based robust NMPC for a tilted multirotor that robustifies the tower-clearance constraint via online constraint tightening. First-order parametric state sensitivities provide a structured-uncertainty margin, while bounded gusts are handled by a stage-dependent additive margin. The formulation augments the nominal NMPC with sensitivity propagation and margin evaluation only, leaving the receding-horizon optimization structure unchanged. Monte-Carlo evaluation over 500 uncertainty realizations on a boundary-critical helical inspection trajectory shows that the proposed controller eliminates the clearance violations observed under nominal NMPC at the cost of a moderate increase in solve time.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

They show that first-order sensitivity propagation plus stage-dependent gust margins can stop clearance violations in Monte-Carlo tests for this tilted-multirotor inspection task, but the letter of the robustness argument is thin.

read the letter

The paper takes nominal NMPC, propagates first-order state sensitivities with respect to parameters like mass and wind, and uses those to tighten the tower-clearance constraint online. It adds a separate stage-dependent additive margin for bounded gusts. The optimization itself stays the same; only the constraint bounds change. On a helical trajectory around the turbine, 500 Monte-Carlo runs with varied mass, inertia, thrust, drag, and wind show zero violations where the nominal controller had some. Solve time goes up modestly. That is the concrete result they deliver for offshore inspection work. The application is narrow but timely, and the implementation looks clean enough that someone could reproduce the controller structure without too much trouble. The Monte-Carlo comparison is the strongest part of the evidence. The soft spot is exactly where the stress-test note points: first-order sensitivities give a linear map, but the clearance constraint is non-convex and the multirotor dynamics are nonlinear. Nothing in the abstract or the reported experiments shows that this linear margin is guaranteed to be an outer approximation when several parameters deviate at once. The 500 samples happened to stay safe, yet that only tests the chosen numbers, not the derivation. No analytic bound or worst-case direction analysis appears to be provided. The uncertainty distributions and how the gust bounds were selected are also left vague. This is useful reading for people who already work on robust NMPC for aerial vehicles or close-proximity drone tasks. A reader who needs a working example of sensitivity-based tightening on a real platform will find the results and the modest overhead worth seeing. It is not a broad methodological advance, but the targeted evidence is clear enough that a serious referee should look at the full derivations and the implementation details. I would send it to review rather than desk-reject.

Referee Report

2 major / 2 minor

Summary. The manuscript proposes a sensitivity-based robust NMPC for a tilted multirotor performing close-proximity offshore wind turbine inspection. It augments nominal NMPC with online first-order parametric state sensitivities to tighten the tower-clearance constraint under uncertainties in mass, inertia, thrust effectiveness, drag, and wind, plus stage-dependent additive margins for bounded gusts. The core receding-horizon optimization remains unchanged. Monte-Carlo evaluation over 500 uncertainty realizations on a boundary-critical helical trajectory shows that the robust controller eliminates all clearance violations observed under nominal NMPC, at the cost of moderate increase in solve time.

Significance. If the first-order margins prove sufficient, the method provides a practical, low-overhead way to robustify existing NMPC formulations for safety-critical UAV tasks without altering the optimization structure or requiring full min-max reformulations. The Monte-Carlo evaluation on a realistic inspection trajectory supplies concrete empirical support for improved constraint satisfaction under combined parametric and gust uncertainty. This could be useful for field deployment in offshore environments where model mismatch is common.

major comments (2)

[§5.2] §5.2 (Monte-Carlo results): The claim that the sensitivity-augmented controller eliminates all clearance violations rests on the unverified premise that first-order parametric sensitivities yield a conservative outer approximation to the reachable set for the nonlinear, non-convex clearance constraint. No comparison to higher-order sensitivities, sampled nonlinear reachable sets, or remainder-term bounds is provided for the combined uncertainty directions (mass/inertia/thrust/drag/wind), so the MC success only demonstrates that the chosen margins worked for the sampled points.
[§4.1] §4.1 (Sensitivity propagation): The structured-uncertainty margin is derived from the linear sensitivity map evaluated along the nominal trajectory. Because the multirotor dynamics and cylindrical clearance constraint are nonlinear, the first-order term can underestimate the true margin in some uncertainty directions; the manuscript provides no a-priori conservatism guarantee or online validation that the linear map remains accurate near the active constraint boundary.

minor comments (2)

[Abstract] Abstract: The description of the method and results is clear but omits any key equations or notation for the sensitivity margin or gust term, forcing readers to consult the body for the technical contribution.
[§3] §3 (Problem formulation): The stage-dependent gust margin bounds are introduced without an explicit formula or tuning procedure; a short derivation or pseudocode would improve reproducibility.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback and for recognizing the practical advantages of the sensitivity-based robust NMPC approach for offshore wind turbine inspection. We address the two major comments point by point below, providing clarifications and indicating where revisions will be made to improve the manuscript.

read point-by-point responses

Referee: [§5.2] §5.2 (Monte-Carlo results): The claim that the sensitivity-augmented controller eliminates all clearance violations rests on the unverified premise that first-order parametric sensitivities yield a conservative outer approximation to the reachable set for the nonlinear, non-convex clearance constraint. No comparison to higher-order sensitivities, sampled nonlinear reachable sets, or remainder-term bounds is provided for the combined uncertainty directions (mass/inertia/thrust/drag/wind), so the MC success only demonstrates that the chosen margins worked for the sampled points.

Authors: We agree that the first-order sensitivity margins constitute an approximation and do not provide a rigorous outer bound on the reachable set for the nonlinear clearance constraint under the combined parametric and gust uncertainties. The Monte-Carlo evaluation with 500 realizations empirically demonstrates that the margins eliminated all observed violations for the sampled uncertainty realizations on the helical trajectory; it does not constitute a proof of conservatism. We will add a dedicated paragraph in §5.2 clarifying the approximate nature of the method, the absence of higher-order or set-propagation comparisons, and the reliance on empirical validation for the considered uncertainty ranges. revision: partial
Referee: [§4.1] §4.1 (Sensitivity propagation): The structured-uncertainty margin is derived from the linear sensitivity map evaluated along the nominal trajectory. Because the multirotor dynamics and cylindrical clearance constraint are nonlinear, the first-order term can underestimate the true margin in some uncertainty directions; the manuscript provides no a-priori conservatism guarantee or online validation that the linear map remains accurate near the active constraint boundary.

Authors: The margin computation in §4.1 is explicitly based on first-order parametric state sensitivities propagated along the nominal trajectory. We acknowledge that nonlinearity in the dynamics and the cylindrical constraint can cause the linear map to underestimate the required margin in certain uncertainty directions, and that no a-priori guarantee or online accuracy check is provided. The online re-computation at each receding-horizon step provides some adaptation, but this does not replace a validation mechanism. We will revise §4.1 to state the first-order linearization assumption explicitly, note the potential for underestimation near active boundaries, and add a short remark on the practical trade-off for the uncertainty magnitudes considered in the paper. revision: yes

Circularity Check

0 steps flagged

No significant circularity; standard sensitivity augmentation with independent MC validation

full rationale

The derivation augments nominal NMPC with first-order parametric state sensitivities for online constraint tightening and stage-dependent additive gust margins. These margins are computed directly from the sensitivity propagation equations applied to the nominal model; they are not fitted to or defined by the Monte-Carlo realizations. The 500-run MC evaluation is presented solely as post-design verification that clearance violations are eliminated, not as input data used to construct the margins or the robust formulation. No self-definitional equations, renamed empirical patterns, or load-bearing self-citations that collapse the central claim back to its own fitted quantities appear in the abstract or described method. The optimization structure remains unchanged, confirming the augmentation is additive rather than circular.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

Abstract-only review limits visibility into exact parameters and assumptions; the approach rests on standard NMPC dynamics and first-order sensitivity validity.

free parameters (1)

gust margin bounds
Stage-dependent additive margins for bounded gusts are introduced; their specific values are not stated in the abstract.

axioms (1)

domain assumption First-order state sensitivities accurately capture the effect of parametric uncertainty on clearance distance
Invoked to justify online constraint tightening without higher-order terms or full uncertainty propagation.

pith-pipeline@v0.9.0 · 5448 in / 1178 out tokens · 47753 ms · 2026-05-11T02:29:46.500348+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

First-order parametric state sensitivities provide a structured-uncertainty margin... Π̇ = ∂f/∂x Π + ∂f/∂ζ, α_p,i ≜ sup |Π_y,i w_ζ| + ε_s
IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

Monte-Carlo evaluation over 500 uncertainty realizations... eliminates the clearance violations observed under nominal NMPC

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

16 extracted references · 16 canonical work pages

[1]

Re- view of robot-based damage assessment for offshore wind turbines,

Y . Liu, M. A. Hajj, R. L. W. van de Werken, and C. G. Soares, “Re- view of robot-based damage assessment for offshore wind turbines,” Renewable and Sustainable Energy Reviews, vol. 158, 2022, art. no. 112187

work page 2022
[2]

Improving Drone Localisation Around Wind Turbines Using Monoc- ular Model-Based Tracking,

O. Moolan-Feroze, K. Karachalios, D. N. Nikolaidis, and A. Calway, “Improving Drone Localisation Around Wind Turbines Using Monoc- ular Model-Based Tracking,” inIEEE International Conference on Robotics and Automation, 2019, pp. 7713–7719

work page 2019
[3]

Au- tonomous Wind-Turbine Blade Inspection Using LiDAR-Equipped Unmanned Aerial Vehicle,

M. Car, L. Markovic, A. Ivanovic, M. Orsag, and S. Bogdan, “Au- tonomous Wind-Turbine Blade Inspection Using LiDAR-Equipped Unmanned Aerial Vehicle,”IEEE Access, vol. 8, pp. 131 380–131 387, 2020

work page 2020
[4]

Non- linear Model Predictive Control with Enhanced Actuator Model for Multi-Rotor Aerial Vehicles with Generic Designs,

D. Bicego, J. Mazzetto, M. Farina, R. Carli, and A. Franchi, “Non- linear Model Predictive Control with Enhanced Actuator Model for Multi-Rotor Aerial Vehicles with Generic Designs,”Journal of Intel- ligent & Robotic Systems, vol. 100, pp. 1213–1247, 2020

work page 2020
[5]

State and Output Feedback Nonlinear Model Predictive Control: An Overview,

R. Findeisen, F. Allg ¨ower, and B. A. Foss, “State and Output Feedback Nonlinear Model Predictive Control: An Overview,”European Journal of Control, vol. 9, no. 2–3, pp. 190–206, 2003

work page 2003
[6]

Model predictive control: Recent developments and future promise,

D. Q. Mayne, “Model predictive control: Recent developments and future promise,”Automatica, vol. 50, no. 12, pp. 2967–2986, 2014

work page 2014
[7]

Multi-stage nonlinear model predictive control applied to a semi-batch polymerization reactor under uncertainty,

S. Lucia, T. Finkler, and S. Engell, “Multi-stage nonlinear model predictive control applied to a semi-batch polymerization reactor under uncertainty,”Journal of Process Control, vol. 23, no. 9, pp. 1306–1319, 2013

work page 2013
[8]

Trajectory Generation for Minimum Closed-Loop State Sensitivity,

P. Robuffo Giordano, Q. Delamare, and A. Franchi, “Trajectory Generation for Minimum Closed-Loop State Sensitivity,” inIEEE International Conference on Robotics and Automation, 2018, pp. 286– 293

work page 2018
[9]

Sensitivity-Aware Model Predictive Control for Robots With Para- metric Uncertainty,

T. Belvedere, M. Cognetti, G. Oriolo, and P. Robuffo Giordano, “Sensitivity-Aware Model Predictive Control for Robots With Para- metric Uncertainty,”IEEE Transactions on Robotics, vol. 41, pp. 3039–3058, 2025

work page 2025
[10]

Zero-order robust nonlinear model predictive control with ellipsoidal uncertainty sets,

A. Zanelli, J. Frey, F. Messerer, and M. Diehl, “Zero-order robust nonlinear model predictive control with ellipsoidal uncertainty sets,” IF AC-PapersOnLine, vol. 54, no. 6, pp. 50–57, 2021

work page 2021
[11]

Stable set- valued integration of nonlinear dynamic systems using affine set- parameterizations,

B. Houska, M. E. Villanueva, and B. Chachuat, “Stable set- valued integration of nonlinear dynamic systems using affine set- parameterizations,”SIAM Journal on Numerical Analysis, vol. 53, no. 5, pp. 2307–2328, 2015

work page 2015
[12]

Design of multirotor aerial vehicles: A taxonomy based on input allocation,

M. Hamandi, F. Usai, Q. Sabl ´e, N. Staub, M. Tognon, and A. Franchi, “Design of multirotor aerial vehicles: A taxonomy based on input allocation,”The International Journal of Robotics Research, vol. 40, no. 8-9, pp. 1015–1044, 2021

work page 2021
[13]

6D interaction control with aerial robots: The flying end-effector paradigm,

M. Ryllet al., “6D interaction control with aerial robots: The flying end-effector paradigm,”The International Journal of Robotics Re- search, vol. 38, no. 9, pp. 1045–1062, 2019

work page 2019
[14]

MATMPC - A MATLAB Based Toolbox for Real-time Nonlinear Model Predictive Control,

Y . Chen, M. Bruschetta, E. Picotti, and A. Beghi, “MATMPC - A MATLAB Based Toolbox for Real-time Nonlinear Model Predictive Control,” inEuropean Control Conference, 2019, pp. 3365–3370

work page 2019
[15]

qpOASES: A parametric active-set algorithm for quadratic program- ming,

H. Ferreau, C. Kirches, A. Potschka, H. Bock, and M. Diehl, “qpOASES: A parametric active-set algorithm for quadratic program- ming,”Mathematical Programming Computation, vol. 6, no. 4, pp. 327–363, 2014

work page 2014
[16]

acados – a modular open-source framework for fast embedded optimal control,

R. Verschueren, G. Frison, D. Kouzoupis, J. Frey, N. van Duijkeren, A. Zanelli, B. Novoselnik, T. Albin, R. Quirynen, and M. Diehl, “acados – a modular open-source framework for fast embedded optimal control,”Mathematical Programming Computation, vol. 14, pp. 147–183, 2021

work page 2021

[1] [1]

Re- view of robot-based damage assessment for offshore wind turbines,

Y . Liu, M. A. Hajj, R. L. W. van de Werken, and C. G. Soares, “Re- view of robot-based damage assessment for offshore wind turbines,” Renewable and Sustainable Energy Reviews, vol. 158, 2022, art. no. 112187

work page 2022

[2] [2]

Improving Drone Localisation Around Wind Turbines Using Monoc- ular Model-Based Tracking,

O. Moolan-Feroze, K. Karachalios, D. N. Nikolaidis, and A. Calway, “Improving Drone Localisation Around Wind Turbines Using Monoc- ular Model-Based Tracking,” inIEEE International Conference on Robotics and Automation, 2019, pp. 7713–7719

work page 2019

[3] [3]

Au- tonomous Wind-Turbine Blade Inspection Using LiDAR-Equipped Unmanned Aerial Vehicle,

M. Car, L. Markovic, A. Ivanovic, M. Orsag, and S. Bogdan, “Au- tonomous Wind-Turbine Blade Inspection Using LiDAR-Equipped Unmanned Aerial Vehicle,”IEEE Access, vol. 8, pp. 131 380–131 387, 2020

work page 2020

[4] [4]

Non- linear Model Predictive Control with Enhanced Actuator Model for Multi-Rotor Aerial Vehicles with Generic Designs,

D. Bicego, J. Mazzetto, M. Farina, R. Carli, and A. Franchi, “Non- linear Model Predictive Control with Enhanced Actuator Model for Multi-Rotor Aerial Vehicles with Generic Designs,”Journal of Intel- ligent & Robotic Systems, vol. 100, pp. 1213–1247, 2020

work page 2020

[5] [5]

State and Output Feedback Nonlinear Model Predictive Control: An Overview,

R. Findeisen, F. Allg ¨ower, and B. A. Foss, “State and Output Feedback Nonlinear Model Predictive Control: An Overview,”European Journal of Control, vol. 9, no. 2–3, pp. 190–206, 2003

work page 2003

[6] [6]

Model predictive control: Recent developments and future promise,

D. Q. Mayne, “Model predictive control: Recent developments and future promise,”Automatica, vol. 50, no. 12, pp. 2967–2986, 2014

work page 2014

[7] [7]

Multi-stage nonlinear model predictive control applied to a semi-batch polymerization reactor under uncertainty,

S. Lucia, T. Finkler, and S. Engell, “Multi-stage nonlinear model predictive control applied to a semi-batch polymerization reactor under uncertainty,”Journal of Process Control, vol. 23, no. 9, pp. 1306–1319, 2013

work page 2013

[8] [8]

Trajectory Generation for Minimum Closed-Loop State Sensitivity,

P. Robuffo Giordano, Q. Delamare, and A. Franchi, “Trajectory Generation for Minimum Closed-Loop State Sensitivity,” inIEEE International Conference on Robotics and Automation, 2018, pp. 286– 293

work page 2018

[9] [9]

Sensitivity-Aware Model Predictive Control for Robots With Para- metric Uncertainty,

T. Belvedere, M. Cognetti, G. Oriolo, and P. Robuffo Giordano, “Sensitivity-Aware Model Predictive Control for Robots With Para- metric Uncertainty,”IEEE Transactions on Robotics, vol. 41, pp. 3039–3058, 2025

work page 2025

[10] [10]

Zero-order robust nonlinear model predictive control with ellipsoidal uncertainty sets,

A. Zanelli, J. Frey, F. Messerer, and M. Diehl, “Zero-order robust nonlinear model predictive control with ellipsoidal uncertainty sets,” IF AC-PapersOnLine, vol. 54, no. 6, pp. 50–57, 2021

work page 2021

[11] [11]

Stable set- valued integration of nonlinear dynamic systems using affine set- parameterizations,

B. Houska, M. E. Villanueva, and B. Chachuat, “Stable set- valued integration of nonlinear dynamic systems using affine set- parameterizations,”SIAM Journal on Numerical Analysis, vol. 53, no. 5, pp. 2307–2328, 2015

work page 2015

[12] [12]

Design of multirotor aerial vehicles: A taxonomy based on input allocation,

M. Hamandi, F. Usai, Q. Sabl ´e, N. Staub, M. Tognon, and A. Franchi, “Design of multirotor aerial vehicles: A taxonomy based on input allocation,”The International Journal of Robotics Research, vol. 40, no. 8-9, pp. 1015–1044, 2021

work page 2021

[13] [13]

6D interaction control with aerial robots: The flying end-effector paradigm,

M. Ryllet al., “6D interaction control with aerial robots: The flying end-effector paradigm,”The International Journal of Robotics Re- search, vol. 38, no. 9, pp. 1045–1062, 2019

work page 2019

[14] [14]

MATMPC - A MATLAB Based Toolbox for Real-time Nonlinear Model Predictive Control,

Y . Chen, M. Bruschetta, E. Picotti, and A. Beghi, “MATMPC - A MATLAB Based Toolbox for Real-time Nonlinear Model Predictive Control,” inEuropean Control Conference, 2019, pp. 3365–3370

work page 2019

[15] [15]

qpOASES: A parametric active-set algorithm for quadratic program- ming,

H. Ferreau, C. Kirches, A. Potschka, H. Bock, and M. Diehl, “qpOASES: A parametric active-set algorithm for quadratic program- ming,”Mathematical Programming Computation, vol. 6, no. 4, pp. 327–363, 2014

work page 2014

[16] [16]

acados – a modular open-source framework for fast embedded optimal control,

R. Verschueren, G. Frison, D. Kouzoupis, J. Frey, N. van Duijkeren, A. Zanelli, B. Novoselnik, T. Albin, R. Quirynen, and M. Diehl, “acados – a modular open-source framework for fast embedded optimal control,”Mathematical Programming Computation, vol. 14, pp. 147–183, 2021

work page 2021