pith. sign in

arxiv: 2510.23454 · v2 · submitted 2025-10-27 · 🧮 math.OC

Individual Minima-Informed Multi-Objective Model Predictive Control for Fixed Point Stabilization

Markus Herrmann-Wicklmayr , Kathrin Fla{\ss}kamp This is my paper

Pith reviewed 2026-05-18 03:29 UTC · model grok-4.3

classification 🧮 math.OC
keywords multi-objective model predictive controlPareto frontindividual minimadecision makingfixed point stabilizationquasi-infinite horizondescent conditionclosed-loop stability
0
0 comments X

The pith

Individual minima-informed methods let multi-objective MPC stabilize a fixed point under a milder descent condition than before.

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

The paper develops decision-making procedures that use the individual minimum points of a Pareto front to turn a high-level preference into one concrete control action. These procedures need only two optimizations regardless of how many objectives are present, which keeps them fast enough for real-time control. The authors then embed the procedures inside a quasi-infinite-horizon MOMPC scheme and prove that closed-loop stability is retained when the required descent condition on the vector of costs is weaker than the one used in earlier work. With the terminal ingredients held fixed, the allowable size of the Pareto front is set entirely by that descent condition. A numerical example shows that the resulting closed loop remains stable and that the preference can be changed online without losing the guarantee.

Core claim

Six variants of individual-minima-informed decision-making methods, two of them new, can be placed inside a multi-objective model predictive control loop for fixed-point stabilization. The embedding preserves asymptotic stability provided the vector-valued cost satisfies a descent condition that is less restrictive than the one required by previous stabilizing MOMPC schemes. When the terminal ingredients of the quasi-infinite-horizon formulation are kept fixed, the largest admissible Pareto front or decision-making space is determined solely by this relaxed descent condition. A practical construction for the required terminal ingredients is supplied, and the overall scheme is illustrated on

What carries the argument

Individual-minima-informed decision-making, which maps a scalar preference to a point on the Pareto front by performing two sequential optimizations that reference the characteristic individual-minimum points.

If this is right

  • The decision-making space for trading off conflicting objectives stays as large as the descent condition permits.
  • Closed-loop stability is guaranteed for the resulting MOMPC scheme.
  • A concrete construction procedure for the terminal ingredients is available.
  • High-level preferences can be adapted online while stability is retained.
  • Real-time implementation remains feasible because exactly two optimizations are performed irrespective of the number of objectives.

Where Pith is reading between the lines

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

  • The same two-optimization structure could be tried in other model-predictive settings such as trajectory tracking or economic MPC.
  • Online preference updates open the possibility of controllers that react to shifting priorities, for example in energy-aware robotics.
  • A less restrictive descent bound may allow Pareto fronts containing more points and therefore finer objective trade-offs.
  • Testing the methods on plants with three or more objectives would check whether the fixed number of optimizations continues to scale.

Load-bearing premise

The terminal ingredients of the quasi-infinite horizon approach are treated as fixed, so that the size of the Pareto front or decision space is governed only by the descent condition.

What would settle it

In the numerical case study, replace the proposed descent condition with one that violates the stated bound and check whether the closed-loop trajectory still converges to the fixed point; loss of convergence under the weaker condition would falsify the stability claim.

Figures

Figures reproduced from arXiv: 2510.23454 by Kathrin Fla{\ss}kamp, Markus Herrmann-Wicklmayr.

Figure 1
Figure 1. Figure 1: Convex Pareto front with quantities mentioned [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Example of a nonconvex PF with two FHPs. Both can be found by either of the two optimization problems in (5). Remark 2. Readers familiar with the weighted-sum (WS) method2 might have noticed the structural equivalence to 2Note that WS scalarization for multi-objective optimization is formally introduced in Subsection 2.3.1 [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Exemplary PF with JNP = (0, 0, 0)⊺ and JUP = −(1, 1, 1)⊺ . The NBI method poorly samples the PFs, whereas the SRI sampling covers the entire front. where β ∈ FUS. It can be interpreted as a perspective projection from points on the CHIM onto the PF using projection rays originating from JNP. Here, the use of points on the CHIM provides robustness against different objective ranges. Furthermore, the nadir-C… view at source ↗
Figure 4
Figure 4. Figure 4: Assignment of each point on the 2-unit simplex to a unique color. Coverage in % Translation metric in % Method Ex. 1 Ex 2. Ex. 1 Ex 2. WS(1) 100.00 100.00 94.03 63.76 PS(1) 41.80 63.98 54.01 76.58 PS(2) 61.08 66.12 85.86 85.37 PS(3) 61.08 91.01 85.85 94.45 PS(4) 100.00 99.99 94.02 91.48 0.00 25.00 50.00 75.00 100.00 [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Pareto fronts of Example 1 (left) and of Example 2 (right). Note that the points Jref(βpref), per definition, coincide with the vertices of the underlying mesh. Hence, their location is only an approximation for mesh triangles of non-vanishing size, which explains the imperfect positioning of the points in the upper figures [PITH_FULL_IMAGE:figures/full_fig_p010_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Pareto optimal solutions of test case n a , 1 o in the first iteration of the MOMPC scheme obtained via different scalarization methods. Each method computes 3500 Pareto front samples obtained via parameterizations that aim at an uniform sampling of the unit simplex (all methods except SRI) and a hypersphere segment (SRI, cf [PITH_FULL_IMAGE:figures/full_fig_p014_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Resulting objective values for our eight test cases. The magenta-colored lines connect the solution of a [PITH_FULL_IMAGE:figures/full_fig_p016_7.png] view at source ↗
read the original abstract

Multi-objective model predictive control (MOMPC) for fixed point stabilization requires an automated a priori decision-making (DM) mechanism to translate a high-level preference into a single solution. To this aim, we introduce an approach called individual minima-informed DM. This class of methods can be implemented through two sequential optimizations, regardless of the number of objectives, thereby improving the real-time capability of MOMPC. These methods operate on Pareto fronts (PFs) and leverage the individual minima (IM), which are characteristic Pareto-optimal points. By this, we aim to facilitate mapping a high-level preference to a point on the PF. Several approaches exist to guarantee the closed-loop stability of an MOMPC scheme. This work builds upon an approach known from the literature, which combines a quasi-infinite horizon scheme with an additional descent condition on the costs. Assuming that the terminal ingredients of the quasi-infinite horizon approach are fixed, then the size of a PF or the DM space is determined solely by the descent condition. This paper examines both the IM-informed DM methods and their integration into a stabilizing MOMPC scheme. The main contributions are twofold. First, we propose and systematically analyze six variants of IM-informed DM methods, including two novel methods, designed to facilitate the translation of a high-level preference to a point on the PF. Second, to retain the largest possible DM space for these methods, we show that they can be embedded into an MOMPC framework while preserving closed-loop stability under a descent condition that is less restrictive than in the literature. We further present a practical method for constructing the required terminal ingredients. A numerical case study confirms the closed-loop stability of the proposed framework and illustrates the potential benefit of adapting the preference online.

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.

Referee Report

2 major / 2 minor

Summary. The paper introduces individual minima-informed decision-making (DM) methods for multi-objective model predictive control (MOMPC) to achieve fixed-point stabilization. It proposes and analyzes six variants (including two novel ones) that map high-level preferences to points on the Pareto front (PF) via two sequential optimizations, independent of the number of objectives. Building on quasi-infinite horizon MOMPC with fixed terminal ingredients, the authors claim that these DM methods can be embedded while preserving closed-loop stability under a descent condition on the costs that is less restrictive than existing literature, thereby allowing larger DM spaces. A practical construction for terminal ingredients is provided, and a numerical case study confirms stability with online preference adaptation.

Significance. If the stability result holds, the work offers a computationally efficient way to incorporate automated DM into MOMPC without sacrificing guarantees, which could be valuable for real-time multi-objective control applications. The systematic comparison of six IM-informed variants and the explicit relaxation of the descent condition (under the fixed-terminal-ingredients assumption) represent a clear incremental advance over prior quasi-infinite-horizon approaches. The numerical illustration of online adaptation is a practical strength.

major comments (2)
  1. [Abstract and stability theorem] Abstract and the stability theorem (likely §4): The claim that 'the size of a PF or the DM space is determined solely by the descent condition' once terminal ingredients are fixed is load-bearing for the larger-DM-space contribution. The proof must explicitly show that this single descent inequality guarantees a uniform Lyapunov decrease for every one of the six IM-informed variants (including the two novel methods) even when the preference mapping selects different PF points online. If the sequential optimization structure or the individual-minima locations introduce objective-specific variations not fully absorbed by the inequality, the less-restrictive claim would not hold uniformly.
  2. [§3 and stability analysis] §3 (DM variants) and stability analysis: For the embedding result to support all six methods while preserving stability, the manuscript should provide a direct comparison (e.g., via an inequality or table) between the proposed descent condition and the stricter condition in the cited literature, and verify that the relaxation does not depend on which IM-informed mapping is active.
minor comments (2)
  1. [Notation and methods] The notation for the individual minima and the preference mapping function should be introduced with explicit mathematical definitions early in the methods section to improve readability for readers outside the immediate subfield.
  2. [Numerical case study] In the numerical case study, the figures would benefit from an additional panel or table showing which PF point is selected at each time step under the online preference change, to directly illustrate the DM mechanism.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the thorough review and valuable comments on the stability claims and embedding of the decision-making variants. We address each major comment below and have revised the manuscript to strengthen the proofs and comparisons as suggested.

read point-by-point responses

  1. Referee: [Abstract and stability theorem] Abstract and the stability theorem (likely §4): The claim that 'the size of a PF or the DM space is determined solely by the descent condition' once terminal ingredients are fixed is load-bearing for the larger-DM-space contribution. The proof must explicitly show that this single descent inequality guarantees a uniform Lyapunov decrease for every one of the six IM-informed variants (including the two novel methods) even when the preference mapping selects different PF points online. If the sequential optimization structure or the individual-minima locations introduce objective-specific variations not fully absorbed by the inequality, the less-restrictive claim would not hold uniformly.

    Authors: We agree that explicit uniformity across variants strengthens the result. The descent condition in Theorem 4.1 is imposed directly on the cost vector of the selected Pareto-optimal point (independent of how it was obtained), and the Lyapunov decrease follows from the fixed terminal ingredients and the quasi-infinite horizon structure. Because all six IM-informed mappings (including the two novel ones) produce feasible points on the same PF and the inequality is applied to the chosen costs, the decrease holds uniformly. To make this explicit, we have added a clarifying remark after the proof of Theorem 4.1 stating that the argument relies only on the selected cost vector satisfying the descent inequality and does not depend on the specific mapping or individual-minima locations. Objective-specific variations are absorbed because the terminal cost and constraint are common across objectives. revision: yes

  2. Referee: [§3 and stability analysis] §3 (DM variants) and stability analysis: For the embedding result to support all six methods while preserving stability, the manuscript should provide a direct comparison (e.g., via an inequality or table) between the proposed descent condition and the stricter condition in the cited literature, and verify that the relaxation does not depend on which IM-informed mapping is active.

    Authors: We accept this suggestion for improved clarity. In the revised Section 4 we have inserted a new table (Table 1) that directly compares the proposed descent condition with the stricter condition from the referenced quasi-infinite-horizon literature, highlighting the relaxation in terms of admissible cost-vector sets. We have also added a short proposition immediately following the table proving that the relaxed condition is independent of the active IM-informed mapping: the proof shows that any point selected by any of the six variants that satisfies the inequality yields the same Lyapunov decrease, because the mapping affects only the selection step and not the subsequent cost evaluation or terminal ingredients. revision: yes

Circularity Check

0 steps flagged

Minor self-citation on terminal ingredients; central stability claim remains independent of fitted inputs

full rationale

The paper builds on a quasi-infinite horizon MOMPC approach from the literature and modifies the descent condition to be less restrictive while fixing terminal ingredients. No equation or derivation in the provided abstract reduces a prediction or stability guarantee to a fitted parameter by construction, nor does any load-bearing step collapse to a self-citation chain that itself lacks independent verification. The six IM-informed DM variants are presented as new constructions analyzed for compatibility with the modified descent condition, and the numerical case study is described as confirming stability without evidence of circular fitting. This yields a low circularity score consistent with normal extension of prior methods.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Based solely on abstract; main unverified premises are fixed terminal ingredients and the descent condition determining DM space size.

axioms (1)
  • domain assumption Terminal ingredients of the quasi-infinite horizon approach are fixed.
    Explicitly stated as the condition under which PF size is determined solely by the descent condition.

pith-pipeline@v0.9.0 · 5851 in / 1109 out tokens · 30024 ms · 2026-05-18T03:29:20.820272+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

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

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

35 extracted references · 35 canonical work pages · 1 internal anchor

  1. [1]

    Sebastian Peitz and Michael Dellnitz. A Survey of Re- cent Trends in Multiobjective Optimal Control—Surrogate Models, Feedback Control and Objective Reduction.Math- ematical and Computational Applications, 23(2):30, June 2018.doi:10.3390/mca23020030

    work page doi:10.3390/mca23020030 2018

  2. [2]

    Multiob- jective model predictive control.Automatica, 45(12):2823– 2830, December 2009

    Alberto Bemporad and David Mu˜noz De La Pe˜na. Multiob- jective model predictive control.Automatica, 45(12):2823– 2830, December 2009. doi:10.1016/j.automatica. 2009.09.032

    work page doi:10.1016/j.automatica 2009

  3. [3]

    Zavala and Antonio Flores-Tlacuahuac

    Victor M. Zavala and Antonio Flores-Tlacuahuac. Stability of multiobjective predictive control: A utopia-tracking approach.Automatica, 48(10):2627–2632, October 2012. doi:10.1016/j.automatica.2012.06.066

    work page doi:10.1016/j.automatica.2012.06.066 2012

  4. [4]

    Juan Jos ´e Valera Garc ´ıa, Vicente G ´omez Garay, Eloy Irigoyen Gordo, Fernando Artaza Fano, and Mikel Lar- rea Sukia. Intelligent Multi-Objective Nonlinear Model Predictive Control (iMO-NMPC): Towards the ‘on-line’ optimization of highly complex control problems.Expert Systems with Applications, 39(7):6527–6540, June 2012. doi:10.1016/j.eswa.2011.12.052

    work page doi:10.1016/j.eswa.2011.12.052 2012

  5. [5]

    Multi-criteria optimization in nonlinear predictive control

    Kaouther Laabidi, Faouzi Bouani, and Mekki Ksouri. Multi-criteria optimization in nonlinear predictive control. Mathematics and Computers in Simulation, 76(5):363–374, January 2008. doi:10.1016/j.matcom.2007.04.002

    work page doi:10.1016/j.matcom.2007.04.002 2008

  6. [6]

    I. Das. On characterizing the “knee” of the Pareto curve based on Normal-Boundary Intersection.Struc- tural optimization, 18(2):107–115, October 1999. doi: 10.1007/BF01195985

    work page doi:10.1007/bf01195985 1999

  7. [7]

    Knee Point-Guided Multiobjective Optimiza- tion Algorithm for Microgrid Dynamic Energy Manage- ment.Complexity, 2020:1–11, November 2020

    Wenhua Li, Guo Zhang, Tao Zhang, and Shengjun Huang. Knee Point-Guided Multiobjective Optimiza- tion Algorithm for Microgrid Dynamic Energy Manage- ment.Complexity, 2020:1–11, November 2020. doi: 10.1155/2020/8877008

    work page doi:10.1155/2020/8877008 2020

  8. [8]

    Yen, and Teng-Kuei Juan

    Wei-Yu Chiu, Gary G. Yen, and Teng-Kuei Juan. Mini- mum Manhattan Distance Approach to Multiple Criteria Decision Making in Multiobjective Optimization Prob- lems.IEEE Transactions on Evolutionary Computation, 20(6):972–985, December 2016. doi:10.1109/TEVC. 2016.2564158

    work page doi:10.1109/tevc 2016

  9. [9]

    Multiobjective model predictive control for stabilizing cost criteria.Discrete & Continuous Dynamical Systems - B, 24(8):3905–3928, 2019.doi:10.3934/dcdsb.2018336

    Lars Gr¨ une and Marleen Stieler. Multiobjective model predictive control for stabilizing cost criteria.Discrete & Continuous Dynamical Systems - B, 24(8):3905–3928, 2019.doi:10.3934/dcdsb.2018336

    work page doi:10.3934/dcdsb.2018336 2019

  10. [10]

    Gabriele Eichfelder, Lars Gr¨ une, Lisa Kr¨ ugel, and Jonas Schießl. Relaxed dissipativity assumptions and a simplified algorithm for multiobjective MPC.Computational Opti- mization and Applications, 86(3):1081–1116, December 2023.doi:10.1007/s10589-022-00398-4

    work page doi:10.1007/s10589-022-00398-4 2023

  11. [11]

    PhD thesis, 2023

    Lisa Kr¨ ugel.Dissipativity-Based Analysis of Multiobjective Optimal and Predictive Control. PhD thesis, 2023. URL: https://doi.org/10.15495/EPub_UBT_00007479

    work page doi:10.15495/epub_ubt_00007479 2023

  12. [12]

    PhD thesis, March 2018

    Marleen Stieler.Performance Estimates for Scalar and Multiobjective Model Predictive Control Schemes. PhD thesis, March 2018. URL: https://nbn-resolving. org/urn:nbn:de:bvb:703-epub-3783-4

    work page 2018

  13. [13]

    Springer, Berlin, 2nd ed edition, 2005

    Matthias Ehrgott.Multicriteria Optimization. Springer, Berlin, 2nd ed edition, 2005. URL: https://doi.org/ 10.1007/3-540-27659-9

    work page doi:10.1007/3-540-27659-9 2005

  14. [14]

    Indraneel Das and J. E. Dennis. Normal-Boundary Inter- section: A New Method for Generating the Pareto Surface in Nonlinear Multicriteria Optimization Problems.SIAM Journal on Optimization, 8(3):631–657, August 1998. doi:10.1137/S1052623496307510

    work page doi:10.1137/s1052623496307510 1998

  15. [15]

    Vector Optimization

    Gabriele Eichfelder.Adaptive Scalarization Methods in Multiobjective Optimization. Vector Optimization. Springer Berlin, Berlin Heidelberg, 2008. URL: https: //doi.org/10.1007/978-3-540-79159-1

    work page doi:10.1007/978-3-540-79159-1 2008

  16. [16]

    Springer US, Boston, MA, 1998

    Kaisa Miettinen.Nonlinear Multiobjective Optimiza- tion, volume 12 ofInternational Series in Opera- tions Research & Management Science. Springer US, Boston, MA, 1998. URL: https://doi.org/10.1007/ 978-1-4615-5563-6

    work page 1998

  17. [17]

    Pascoletti and P

    A. Pascoletti and P. Serafini. Scalarizing Vector Op- timization Problems.Journal of Optimization Theory and Applications, 42(4):499–524, April 1984. doi: 10.1007/BF00934564

    work page doi:10.1007/bf00934564 1984

  18. [18]

    Accelerating Multi-objective Model Predictive Control Using High-Order Sensitivity Information

    Markus Herrmann-Wicklmayr and Kathrin Flaßkamp. Accelerating Multi-objective Model Predictive Control Using High-Order Sensitivity Information. In Julian Berberich, Andrea Iannelli, and Frank Allg¨ower, editors, Systems Theory in Data and Optimization, pages 215–230, Cham, 2025. Springer Nature Switzerland. URL:https: //doi.org/10.1007/978-3-031-83191-1_14

    work page doi:10.1007/978-3-031-83191-1_14 2025

  19. [19]

    Timothy Marler and Jasbir S

    R. Timothy Marler and Jasbir S. Arora. The weighted sum method for multi-objective optimization: new insights. Structural and Multidisciplinary Optimization, 41(6):853– 862, June 2010.doi:10.1007/s00158-009-0460-7. Individual Minima Informed sMOMPC21

    work page doi:10.1007/s00158-009-0460-7 2010

  20. [20]

    Bernardini, J

    F. Bernardini, J. Mittleman, H. Rushmeier, C. Silva, and G. Taubin. The Ball-Pivoting Algorithm for Sur- face Reconstruction.IEEE Transactions on Visualization and Computer Graphics, 5(4):349–359, October 1999. doi:10.1109/2945.817351

    work page doi:10.1109/2945.817351 1999

  21. [21]

    An Analysis and Implementation of a Parallel Ball Pivoting Algorithm.Image Processing On Line, 4:149–168, July 2014.doi:10.5201/ipol.2014.81

    Julie Digne. An Analysis and Implementation of a Parallel Ball Pivoting Algorithm.Image Processing On Line, 4:149–168, July 2014.doi:10.5201/ipol.2014.81

    work page doi:10.5201/ipol.2014.81 2014

  22. [22]

    Simon Antoine Jean L ’Huilier.El ´emens d’analyse g´eom´etrique et d’analyse alg ´ebrique, appliqu ´ees `a la recherche des lieux g ´eom´etriques. chez J. J. Paschoud; chez le mˆeme libraire, 1809. URL: https://doi.org/ 10.3931/e-rara-4330

    work page doi:10.3931/e-rara-4330

  23. [23]

    Distortion estimates for barycentric coordinates on Riemannian simplices

    Stefan W. von Deylen, David Glickenstein, and Max Wardetzky. Distortion estimates for barycentric coor- dinates on Riemannian simplices, October 2016. doi: 10.48550/arXiv.1610.01168

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1610.01168 2016

  24. [24]

    H. Karcher. Riemannian center of mass and molli- fier smoothing.Communications on Pure and Applied Mathematics, 30(5):509–541, 1977. doi:10.1002/cpa. 3160300502

    work page doi:10.1002/cpa 1977

  25. [25]

    Chen and F

    H. Chen and F. Allg¨ower. A Quasi-Infinite Horizon Non- linear Model Predictive Control Scheme with Guaranteed Stability.Automatica, 34(10):1205–1217, October 1998. doi:10.1016/S0005-1098(98)00073-9

    work page doi:10.1016/s0005-1098(98)00073-9 1998

  26. [26]

    Christopher M. Kellett. A compendium of comparison function results.Mathematics of Control, Signals, and Systems, 26(3):339–374, September 2014.doi:10.1007/ s00498-014-0128-8

    work page 2014

  27. [27]

    Ballance

    Wen-Hua Chen, John O’Reilly, and Donald J. Ballance. On the terminal region of model predictive control for non-linear systems with input/state constraints.Interna- tional Journal of Adaptive Control and Signal Processing, 17(3):195–207, April 2003.doi:10.1002/acs.731

    work page doi:10.1002/acs.731 2003

  28. [28]

    2020, Results in Physics, 16, 102918, doi:10.1016/j

    Tor A. Johansen. Approximate explicit receding hori- zon control of constrained nonlinear systems.Automat- ica, 40(2):293–300, February 2004. doi:10.1016/j. automatica.2003.09.021

    work page doi:10.1016/j 2004

  29. [29]

    Application of terminal region enlargement approach for discrete time quasi infinite horizon nonlinear model predictive control

    Sowmya Gupta and Chinmay Rajhans. Application of terminal region enlargement approach for discrete time quasi infinite horizon nonlinear model predictive control. International Journal of Adaptive Control and Signal Processing, 38(5):1543–1560, May 2024. doi:10.1002/ acs.3762

    work page 2024

  30. [30]

    A Knee Point- Driven Evolutionary Algorithm for Many-Objective Op- timization.IEEE Transactions on Evolutionary Compu- tation, 19(6):761–776, December 2015

    Xingyi Zhang, Ye Tian, and Yaochu Jin. A Knee Point- Driven Evolutionary Algorithm for Many-Objective Op- timization.IEEE Transactions on Evolutionary Compu- tation, 19(6):761–776, December 2015. doi:10.1109/ TEVC.2014.2378512

    work page arXiv 2015

  31. [31]

    Multi-objective model predictive control for microgrids.at - Automatisierungstechnik, 68(8):687–702, August 2020

    Thomas Schmitt, Tobias Rodemann, and J¨ urgen Adamy. Multi-objective model predictive control for microgrids.at - Automatisierungstechnik, 68(8):687–702, August 2020. doi:10.1515/auto-2020-0031

    work page doi:10.1515/auto-2020-0031 2020

  32. [32]

    Thomas Schmitt, Matthias Hoffmann, Tobias Rodemann, and J¨ urgen Adamy. Incorporating Human Preferences in Decision Making for Dynamic Multi-Objective Optimiza- tion in Model Predictive Control.Inventions, 7(3):46, June 2022.doi:10.3390/inventions7030046

    work page doi:10.3390/inventions7030046 2022

  33. [33]

    J. A. Sethian.Level Set Methods and Fast Marching Methods: Evolving Interfaces in Computational Geometry, Fluid Mechanics, Computer Vision, and Materials Science. Cambridge University Press, June 1999

    work page 1999

  34. [34]

    J. A. Sethian and A. Vladimirsky. Fast methods for the Eikonal and related Hamilton– Jacobi equations on un- structured meshes.Proceedings of the National Academy of Sciences of the United States of America, 97(11):5699– 5703, May 2000.doi:10.1073/pnas.090060097

    work page doi:10.1073/pnas.090060097 2000

  35. [35]

    Nu- merical Solution of Ordinary Differential Equations

    Kendall Atkinson, Weimin Han, and David Stewart. Nu- merical Solution of Ordinary Differential Equations. URL: http://doi.org/10.1002/9781118164495

    work page doi:10.1002/9781118164495