Pith. sign in

REVIEW 3 major objections 4 minor 34 references

Dual Hierarchical Least-Squares Programming with Equality Constraints

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read The paper establishes that equality-constrained hierarchical least-squares programs can be reformulated as a single convex, differentiable quadratically constrained least-squares program, and gives an ADMM solver that is about an order of…

desk verdict The ADMM solver and gradient are real contributions, but the central equivalence claim is unproven and the paper's own experiments suggest the simplified dual is a relaxation, not an exact reformulation. read the letter →

arxiv 2505.21071 v2 pith:VIRYVP5M submitted 2025-05-27 math.OC

Kai Pfeiffer This is my paper
classification math.OC MSC 90C2590C2090C46
keywords hierarchicalleast-squaresprogrammingdualformulationquadraticallyconstrainedADMMequalityconstraintsdifferentiabilityinterior-pointmethod
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

Hierarchical least-squares programming (HLSP) resolves stacked priority levels one after another, which is fast but makes the overall mapping non-differentiable. This paper tries to establish that, for equality constraints, the whole hierarchy can be re-expressed as one convex and differentiable quadratically constrained least-squares program (D-HLSP-E), so all priority levels are handled at once. That would matter because a differentiable HLSP can be embedded in neural networks and solved in distributed settings. The paper also develops an ADMM solver, D-HADM, and reports it is about one order of magnitude faster than the interior-point dual solver D-HIPM, though still slower than classical primal hierarchical solvers.

What carries the argument

The central object is D-HLSP-E: the dual formulation of an equality-constrained HLSP written as a convex quadratically constrained least-squares program. Its constraints are (i) the primal level equations $A_{E_l} x - b_{E_l} = v_{E_l}$, (ii) dual orthogonality conditions $A_{E_l}^T v_{E_l} + A_{E_{\cup l-1}}^T \lambda_{l,E_{\cup l-1}} = 0$, and (iii) convex quadratic dual constraints $(v_{E_l}+\hat b_{E_l})^T(v_{E_l}+\hat b_{E_l}) \le \hat b_{E_l}^T\hat b_{E_l} - \lambda^T b$. The mechanism that makes the solver fast is ADMM operator splitting: split variables $z$ and $\tilde\lambda$ decouple the quadratic constraints and the primal-dual variables, so the substituted KKT system $K^\rho_x x = -k^\rho_x$ factorizes only in $x$, with each quadratic projection reduced to solving a cubic polynomial.

What would settle it

Compare the optimal level-wise slack norms produced by D-HLSP-E with those from a sequential primal HLSP solver on randomly generated equality-constrained hierarchies with linearly dependent constraint matrices; any instance where the two values diverge by more than solver tolerance would disprove the claimed equivalence.

Watch

Extended reading notes

Core claim

The paper's central claim is that every equality-constrained HLSP is equivalent to a single convex QCLSP in which the primal variables, slack variables, and primal-dual variables (the dual variables of higher levels promoted to primal variables) appear together, and in which strong duality holds level by level. The paper supports this by deriving D-HLSP-E from the Lagrangian of a lexicographic hierarchy and proving (Theorem 1) that the duality gap can be closed at each level using free multipliers. It then shows that an ADMM implementation, D-HADM, can substitute the primal-dual variables out of the KKT system so that only a matrix on the primal variable $x$ needs factorization, and reports roughly an order-of-magnitude speedup over the interior-point dual solver D-HIPM on randomized hierarchies with up to ten priority levels and rank-deficient constraints.

Load-bearing premise

The proof assumes the free $\theta$ multipliers can always be chosen nonnegative to force strong-duality equalities level by level, and that dropping all $\theta$ terms afterwards does not change the optimal value of the convex problem; the paper does not rigorously prove that last step.

Editorial extensions

If this is right

  • D-HLSP-E turns a lexicographic hierarchy into one convex QCLSP, so all priority levels are solved simultaneously rather than sequentially.
  • Because the dual program is differentiable, gradients with respect to problem data can be computed by solving a linear system, enabling HLSPs as layers in neural networks.
  • The ADMM formulation keeps matrix factorizations on the primal variable $x$ only, so the cost per iteration does not grow cubically with the number of primal-dual variables.
  • D-HADM is about one order of magnitude faster than the dual interior-point solver D-HIPM on the tested hierarchies with up to ten priority levels.
  • The dual formulation currently applies only to equality-constrained HLSPs; inequality constraints introduce non-convex complementary constraints that remain open.

Reading between the lines

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

  • Editorial inference: A natural testable extension is to verify whether the omitted $\theta$ terms can change the optimal value in infeasible or rank-deficient cases; this would settle whether the convex QCLSP is exactly equivalent to the original hierarchy.
  • Editorial inference: The same operator-splitting idea could be applied to the dual of sparse hierarchical programs with $\ell^0$/ $\ell^1$ norms, where a linear dual already exists; the ADMM projection step would need a different scalar root-finding problem.
  • Editorial inference: The gradient linear system used in the appendix may be solvable by reusing factorizations from D-HADM, which would make neural-network integration computationally practical; the paper leaves that efficiency question open.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. This paper develops a dual formulation of lexicographic least-squares programming with equality constraints (D-HLSP-E), claiming that it is a single convex quadratically constrained least-squares program (QCLSP) equivalent to the sequential primal HLSP. The derivation in Sec. 4 builds a Lagrangian over all priority levels, extracts stationarity and duality-gap conditions, and then simplifies the problem by dropping the third-order terms involving the multipliers θ. Section 5 presents an ADMM solver (D-HADM) whose operator splitting eliminates the primal-dual variables from the matrix factorization, and App. B gives an interior-point counterpart (D-HIPM). The experiments in Sec. 7 compare D-HADM, D-HIPM, and the author's prior primal solver NIPM on randomized and rank-deficient problems, reporting that D-HADM is about one order of magnitude faster than D-HIPM while remaining 2–3 orders of magnitude slower than NIPM, with moderate accuracy. The paper also sketches a gradient computation in App. A as a basis for future differentiable and distributed applications.

Significance. If the claimed equivalence were rigorously established, the contribution would be significant: a differentiable, single-shot convex formulation of hierarchical least-squares would enable new modes of use (neural-network layers, distributed optimization), and the ADMM complexity reduction—factorizing only an n_x × n_x system instead of one including all primal-dual variables—is a genuine algorithmic idea. The paper is transparent about its limitations: Sec. 8 acknowledges the inability to handle inequality constraints and states that the dual solver is slower than primal solvers. The explicit gradient system in App. A and the honest comparison against the author's own solvers are also strengths. However, the equivalence claim is the load-bearing part of the paper, and it is not proven as written; the significance is therefore conditional on a substantial revision of the proof and on reconciling the experimental discrepancy reported in Sec. 7.

major comments (3)
  1. [Sec. 4, Theorem 1] The proof of Theorem 1 does not establish the existence of nonnegative multipliers θ_{l,k} forcing the strong-duality equalities. Substituting the KKT identities into (12) shows that the claimed equality at level l requires Σ_{k<l} θ_{l,k} ||v*_{Ck}||² = λ^T_{l,C∪l−1} v*_{C∪l−1}, and the proof never verifies that the right-hand side is nonnegative or attainable with θ ≥ 0; the sentence that θ* is "a free variable and has been chosen appropriately" is exactly the assertion that needs proof. Moreover, g_p in (10) is the Lagrangian evaluated at a KKT point rather than an infimal dual function, so the weak-duality inequality invoked after (11) is not standard and is not proven. Since (12) is the basis for the duality constraint (16), Theorem 1 as stated does not support the claim that D-HLSP-E enforces zero duality gap at each priority level.
  2. [Sec. 4, Eq. (16), convex simplification] The transition from the full duality constraint (16), including the third-order term Σ_{k<l} θ_{l,k} v^T_{Ck} v_{Ck}, to the implemented QCLSP in which all θ terms are dropped is asserted rather than derived. The argument that θ_{l,k} is uncoupled and can be chosen freely only shows that θ acts as a slack variable in the full formulation; it does not show that the θ-free inequality (16) has the same feasible set or the same optimal value as the full constraint, nor that it remains tight enough to enforce the lexicographic optimum. The claim that θ = 0 favors strong duality in (12) is also not backed by an inequality analysis, since strong duality may require specific nonzero θ values or may be unattainable with θ ≥ 0. Consequently the abstract's assertion that the simplified D-HLSP-E is a convex QCLSP equivalent to the lexicographic HLSP is not established; at minimum, the authors need to prove that every minimizer of the θ-free problem attains the lexicographic optimal slack values at all higher-priority levels.
  3. [Sec. 7, Fig. 2] The rank-deficient experiment in Fig. 2 shows the dual solvers reporting ||v_{E5}||² = 106.7 while the primal solver NIPM reports 110.5. If the claimed equivalence held, the lexicographic optimal value at level 5 would be unique and both solvers should approach it; the text's statement that the dual values are "more accurate" has no stated ground truth, and it sits in tension both with Theorem 1 and with Table 1, which classifies D-HADM as "less accurate." The authors should resolve this by reporting the level-by-level lexicographic subproblem values as a ground truth, or by attributing the gap to convergence tolerances and conditioning of the 1e-12-perturbed rank-deficient matrices with explicit evidence. As written, the experiment is consistent with the possibility that the simplified QCLSP is a relaxation, which would contradict the central claim.
minor comments (4)
  1. [Sec. 5.3 and Sec. 7] The subscript in "with 2 = 1, . . . , p−1" following Eq. (41) should read l = 1, and "D-HADM aglorithm 1" in Sec. 7 should read "Algorithm 1."
  2. [Sec. 7, experimental setup] The experiments do not report the values of the stopping tolerance χ, the over-relaxation parameter α, the ρ-update parameters, or the stopping criteria used for D-HIPM; these should be specified for reproducibility, since the accuracy claims ("moderate at levels of 1e-2") depend on them.
  3. [App. A, gradient] The differentiability claim is only supported by the linear system (55); differentiability of the solution map also requires conditions such as uniqueness of the KKT solution and a constraint qualification, which are not discussed. The QCLSP's inequality constraints can become active or inactive, and the authors should state the assumptions under which the gradient system is valid.
  4. [Table 1 and Sec. 7] The application column of Table 1 lists D-HADM as "less accurate," which is in tension with the Sec. 7 characterization of the dual solvers' values as "more accurate" in the rank-deficient sample; the relationship between these two statements should be clarified in the text.

Circularity Check

1 steps flagged · score 6.0 of 10

The convex QCLSP's claimed equivalence to lexicographic HLSP is purchased by choosing the θ multipliers to force strong duality in Theorem 1, then discarding those θ terms in the problem actually solved.

  1. self definitional [Sec. 4, Theorem 1 proof (around Eq. 16) and the simplification following Eq. 19]
    "θ∗ 2,1 ≥ 0 is a free variable and has been chosen appropriately such that the above equality holds. Following this pattern for all priority levels 1 to p, strong duality ... follows for D-HLSP-E. ... With the above simplifications, D-HLSP-E becomes a convex QCLSP as all colored terms are ignored."

    Theorem 1 is meant to justify that the convex QCLSP D-HLSP-E is exactly equivalent to the equality-constrained lexicographic HLSP. But its proof establishes strong duality only for the problem that still contains the free θ terms, by choosing θ to force the strong-duality equality (16) to hold. The D-HLSP-E that is actually formulated and solved in Sec. 5 is obtained by setting all θ terms to zero ("all colored terms are ignored"). Thus the theorem proves a statement about a different problem: it assumes the existence of θ making (16) tight, then the solver drops θ, leaving (16) as a possibly strict inequality. The claimed equivalence is therefore not derived; it is built into the proof by selecting θ to make the conclusion true and then discarding that selection. The paper's own Sec.

full rationale

The QCLSP formulation, the ADMM operator splitting, and the reduced KKT system in Sec. 5 are derived from the paper's own equations and are not fitted to any benchmark; the speed comparison D-HADM versus D-HIPM is an implementation comparison rather than a circular derivation. The single load-bearing circular move is in Sec. 4: Theorem 1 claims strong duality for the equality-constrained D-HLSP-E, but its proof selects the free multipliers θ to make the strong-duality equality hold, while the convex QCLSP analyzed and solved in Sec. 5 is obtained by setting all θ terms to zero. Consequently, the theorem applies to a different, θ-dependent problem, and the equivalence of the simplified QCLSP to Lex-HLSP is assumed rather than proven. The paper's own Fig. 2 (§7) reports different level-5 objectives for the dual and primal solvers, which contradicts the claimed equivalence unless the dual program is a relaxation. Self-citations in refs. [25]–[29] are used as prior solvers and benchmarks, but none of them carries a uniqueness or ansatz argument that makes the central derivation depend on unverifiable self-support. The score reflects the partial circularity in the central equivalence claim, not the ADMM engineering, which is otherwise self-contained.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central claim rests on a convex QCLSP obtained by dropping non-convex terms; the paper provides a proof sketch for the equivalence, but the proof is incomplete and the simplification is not independently validated. The free parameters are mostly algorithmic tuning constants rather than data-fitted quantities.

free parameters (4)
  • ADMM step-size ratios rho_mu, rho_eta, rho_phi, rho_nu = 100, 10, 1, 1
    Chosen by hand in Sec. 5.1 to bias updates; not derived from the problem.
  • regularization sigma = greater than 0
    Introduced to make the KKT system positive-definite; the paper does not specify its value or update rule.
  • over-relaxation alpha = in (0,2), no specific value
    Mentioned in Sec. 5.7 as an enhancement, but no value is given in the algorithm.
  • stopping tolerance chi = small
    Required for the convergence check in Algorithm 1, but no numerical value is provided.
assumptions (4)
  • standard math Strong duality holds for equality-constrained QPs with affine constraints.
    Used in the proof of Theorem 1 as a starting point for each priority level.
  • ad hoc to paper The multipliers theta can be chosen nonnegative such that the strong-duality equalities hold for every priority level.
    This is the core of the Theorem 1 proof; it is asserted but not rigorously proven, and later the paper drops all theta terms to obtain convexity.
  • ad hoc to paper Dropping the positive third-order term in (16) and the complementary constraint (18) preserves the optimal solution for equality-only hierarchies.
    This simplification turns D-HLSP-E into a convex QCLSP, but the paper does not prove that the nominal lexicographic solution remains optimal after the simplification.
  • domain assumption The ADMM converges to the global minimizer of the convex QCLSP.
    Standard for convex problems, but requires constraint qualification (e.g., Slater's condition), which is not discussed in the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Dual Hierarchical Least-Squares Programming with Equality Constraints." pith.science (2026). https://pith.science/paper/VIRYVP5M

@misc{pith2026250521071,
  author       = {Pith},
  title        = {Pith review of: Dual Hierarchical Least-Squares Programming with Equality Constraints},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VIRYVP5M}},
  note         = {Machine review of arXiv:2505.21071}
}
read the original abstract

Hierarchical least-squares programming (HLSP) is an important tool in optimization as it enables the stacking of any number of priority levels in order to reflect complex constraint relationships, for example in physical systems like robots. Existing solvers typically address the primal formulation of HLSP's, which is computationally efficient due to sequential treatment of the priority levels. This way, already identified active constraints can be eliminated after each priority level, leading to smaller problems as the solver progresses through the hierarchy. However, this sequential progression makes the solvers discontinuous and therefore not differentiable. This prevents the incorporation of HLSP's as neural network neurons, or solving HLSP's in a distributed fashion. In this work, an efficient solver based on the dual formulation of HLSP's with equality constraints (D-HLSP-E) is developed. D-HLSP-E is a convex and differentiable quadratically constrained least-squares program (QCLSP), which is solved by an Alternating Direction Method of Multipliers (ADMM). By introducing appropriate operator splitting, primal-dual variables, which link each priority level with all their respective higher priority levels, can be eliminated from the main computation step of computing a matrix factorization. The proposed solver D-HADM is about one magnitude faster than a comparable D-HLSP-E solver based on the interior-point method (D-HIPM), where the primal-dual variables enter the computational complexity in a cubic fashion.

Figures

Figures reproduced from arXiv: 2505.21071 by the authors.

Figure 1
Figure 1. Solver data of D-HADM, D-HIPM, and N IPM [27] for hierarchies of p = 1, . . . , 10 levels with equality constraints. Shown from top to bottom are the solver times, norm of the KKT residuals, number of solver iterations, and the number of primal￾dual variables |Λp−1|. Comparison of the behavior of the different HLSP solvers D-HADM (D-HLSP-E solver, based on ADMM), D-HIPM (D-HLSP-E solver, based on IPM), and N IPM [27… view at source ↗
Figure 2
Figure 2. Primal variables x and norm of vEl for an instance of a test with p = n = 10, in the case of full rank (left) and rank￾deficiency (right) of each constraint matrix AEl (with l = 1, . . . , p). primal-dual variables |Λp−1|. For example, for p = 10, D-HADM is about one magnitude faster than D-HIPM. For an instance of solving a problem with p = 9, individual parts of the D-HADM aglorithm 1 make up following parts of th… view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

34 extracted references · 32 canonical work pages

  1. [30]

    A weighted method for fast resolution of strictly hi- erarchical robot task specifications using exact penalty functions

    Ajay Suresha Sathya, Goele Pipeleers, Wilm Decr ´e, and Jan Swevers. A weighted method for fast resolution of strictly hi- erarchical robot task specifications using exact penalty functions. IEEE Robotics and Automation Letters , 6(2):3057–3064, 2021

    work page 2021

  2. [1]

    Zico Kolter

    Brandon Amos and J. Zico Kolter. OptNet: Differentiable optimization as a layer in neural networks. In Proceedings of the 34th International Conference on Machine Learning, volume 70 ofProceedings of Machine Learning Research, pages 136–145. PMLR, 2017

    2017

  3. [2]

    Anstreicher

    Kurt M. Anstreicher. On convex relaxations for quadratically constrained quadratic programming. Mathematical Programming, 136(2):233–251, Dec 2012

    work page 2012

  4. [3]

    Distributed optimization and statistical learning via the alternating direction method of multipliers

    Stephen Boyd, Neal Parikh, Eric Chu, Borja Peleato, and Jonathan Eckstein. Distributed optimization and statistical learning via the alternating direction method of multipliers. F oundations and Trends in Machine Learning, 3:1–122, 01 2011

    work page 2011

  5. [4]

    Convex Optimization

    Stephen Boyd and Lieven Vandenberghe. Convex Optimization. Cambridge University Press, March 2004

    work page 2004

  6. [5]

    A sublevel moment-sos hierarchy for polynomial optimization

    Tong Chen, Jean Bernard Lasserre, Victor Magron, and Edouard Pauwels. A sublevel moment-sos hierarchy for polynomial optimization. Computational Optimization and Applications , 81:31 – 66, 2021

    work page 2021

  7. [6]

    A hierarchical framework for collision avoidance in robot-assisted minimally invasive surgery

    Jacinto Colan, Ana Davila, Khusniddin Fozilov, and Yasuhisa Hasegawa. A hierarchical framework for collision avoidance in robot-assisted minimally invasive surgery. In2024 IEEE International Conference on Cyborg and Bionic Systems (CBS) , pages 133–138, 2024

    work page 2024

  8. [7]

    Dang, K.V

    Thuy V . Dang, K.V . Ling, and J.M. Maciejowski. Embedded admm-based qp solver for mpc with polytopic constraints. In 2015 European Control Conference (ECC), pages 3446–3451, 2015

    work page 2015

Show all 34 references
  1. [8]

    Efficient resolution of potentially conflicting linear constraints in robotics

    Dimitar Dimitrov, Alexander Sherikov, and Pierre-Brice Wieber. Efficient resolution of potentially conflicting linear constraints in robotics. August 2015

    2015

  2. [9]

    Hierarchical quadratic programming: Fast online humanoid-robot motion generation

    Adrien Escande, Nicolas Mansard, and Pierre-Brice Wieber. Hierarchical quadratic programming: Fast online humanoid-robot motion generation. The International Journal of Robotics Research , 33(7):1006–1028, 2014

    2014

  3. [10]

    Roger Fletcher, Sven Leyffer, and Philippe L. Toint. On the global convergence of a filter–sqp algorithm. SIAM Journal on Optimization, 13(1):44–59, 2002

    2002

  4. [11]

    Optimal parameter selection for the alternating direction method of multipliers (admm): Quadratic problems

    Euhanna Ghadimi, Andr ´e Teixeira, Iman Shames, and Mikael Johansson. Optimal parameter selection for the alternating direction method of multipliers (admm): Quadratic problems. IEEE Transactions on Automatic Control, 60(3):644–658, 2015

    2015

  5. [12]

    Gill, Walter Murray, Michael A

    Philip E. Gill, Walter Murray, Michael A. Saunders, and Margaret H. Wright. A practical anti-cycling procedure for linearly constrained optimization. Mathematical Programming, 45(1):437–474, Aug 1989

    1989

  6. [13]

    Task priority based design optimization of a kine- matic redundant robot

    Angelica Ginnante, Enrico Simetti, St ´ephane Caro, and Franc ¸ois Leborne. Task priority based design optimization of a kine- matic redundant robot. Mechanism and Machine Theory, 187:105374, 2023

    2023

  7. [14]

    Gurobi Optimizer Reference Manual, 2023

    Gurobi Optimization, LLC. Gurobi Optimizer Reference Manual, 2023

    2023

  8. [15]

    Lexicographic mpc with multiple economic criteria for constrained nonlinear systems

    Defeng He, Shiming Yu, and Linlin Ou. Lexicographic mpc with multiple economic criteria for constrained nonlinear systems. Journal of the Franklin Institute, 355(2):753–773, 2018

    2018

  9. [16]

    Momentum control with hierarchical inverse dynamics on a torque-controlled humanoid

    Alexander Herzog, Nicholas Rotella, Sean Mason, Felix Grimminger, Stefan@misc Schaal, and Ludovic Righetti. Momentum control with hierarchical inverse dynamics on a torque-controlled humanoid. Autonomous Robots, 40(3):473–491, Mar 2016

    2016

  10. [17]

    O Kanoun, F Lamiraux, and P-B. Wieber. Kinematic control of redundant manipulators: generalizing the task priority frame- work to inequality tasks. IEEE Trans. on Robotics, 27(4):785–792, 2011

    2011

  11. [18]

    Knight, Daniel Ruiz, and Bora Uc ¸ar

    Philip A. Knight, Daniel Ruiz, and Bora Uc ¸ar. A symmetry preserving algorithm for matrix scaling. SIAM Journal on Matrix Analysis and Applications, 35(3):931–955, 2014

    2014

  12. [19]

    Cooperative multi-robot object transportation system based on hierarchical quadratic programming

    Daravuth Koung, Olivier Kermorgant, Isabelle Fantoni, and Lamia Belouaer. Cooperative multi-robot object transportation system based on hierarchical quadratic programming. IEEE Robotics and Automation Letters , 6(4):6466–6472, 2021

    2021

  13. [20]

    A sparse admm-based solver for linear mpc subject to terminal quadratic constraint

    Pablo Krupa, Rim Jaouani, Daniel Limon, and Teodoro Alamo. A sparse admm-based solver for linear mpc subject to terminal quadratic constraint. IEEE Transactions on Control Systems Technology, 32(6):2376–2384, 2024

    2024

  14. [21]

    An efficiently solvable quadratic program for stabilizing dynamic locomotion

    Scott Kuindersma, Frank Permenter, and Russ Tedrake. An efficiently solvable quadratic program for stabilizing dynamic locomotion. Proceedings - IEEE International Conference on Robotics and Automation , pages 2589–2594, 2014

    2014

  15. [22]

    Differentiable simulation for physical system identification

    Quentin Le Lidec, Igor Kalevatykh, Ivan Laptev, Cordelia Schmid, and Justin Carpentier. Differentiable simulation for physical system identification. IEEE Robotics and Automation Letters , 6(2):3413–3420, 2021

    2021

  16. [23]

    Optimal parameter selection for admm: Quadratically constrained quadratic program

    H.-N Nguyen. Optimal parameter selection for admm: Quadratically constrained quadratic program. IEEE Transactions on Automatic Control, PP:1–16, 01 2025

    2025

  17. [24]

    Neal Parikh and Stephen Boyd. 2014

    2014

  18. [25]

    Sparse hierarchical non-linear programming for inverse kinematic planning and control with autonomous goal selection

    Kai Pfeiffer. Sparse hierarchical non-linear programming for inverse kinematic planning and control with autonomous goal selection. https://arxiv.org/abs/2412.01324, 2024

    2024 arXiv

  19. [26]

    The hierarchical newton’s method for numerically stable prioritized dynamic control

    Kai Pfeiffer, Adrien Escande, Pierre Gergondet, and Abderrahmane Kheddar. The hierarchical newton’s method for numerically stable prioritized dynamic control. IEEE Transactions on Control Systems Technology, pages 1–14, 2023

    2023

  20. [27]

    N ipm-hlsp: an efficient interior-point method for hierarchical least-squares programs

    Kai Pfeiffer, Adrien Escande, and Ludovic Righetti. N ipm-hlsp: an efficient interior-point method for hierarchical least-squares programs. Optimization and Engineering, 25(2):759–794, June 2024

    2024

  21. [28]

    Efficient lexicographic optimization for prioritized robot control and planning

    Kai Pfeiffer and Abderrahmane Kheddar. Efficient lexicographic optimization for prioritized robot control and planning. https://arxiv.org/abs/2403.09160, 2024

    2024 arXiv

  22. [29]

    Sequential hierarchical least-squares programming for prioritized non-linear optimal control

    Kai Pfeiffer and Abderrahmane Kheddar. Sequential hierarchical least-squares programming for prioritized non-linear optimal control. Optimization Methods and Software, 39(5):1104–1142, 2024

    2024

  23. [31]

    Sherali and Allen L

    Hanif D. Sherali and Allen L. Soyster. Preemptive and nonpreemptive multi-objective programming: Relationship and coun- terexamples. Journal of Optimization Theory and Applications , 39:173–186, 1983

    1983

  24. [32]

    Stellato, G

    B. Stellato, G. Banjac, P. Goulart, A. Bemporad, and S. Boyd. OSQP: an operator splitting solver for quadratic programs. Mathematical Programming Computation, 12(4):637–672, 2020

    2020

  25. [33]

    Multi- contact vertical ladder climbing with an hrp-2 humanoid

    Joris Vaillant, Abderrahmane Kheddar, Herv ´e Audren, Franc ¸ois Keith, Stanislas Brossette, Adrien Escande, Karim Bou- yarmane, Kenji Kaneko, Mitsuharu Morisawa, Pierre Gergondet, Eiichi Yoshida, Suuji Kajita, and Fumio Kanehiro. Multi- contact vertical ladder climbing with a...

    2016

  26. [34]

    Hqp-based obstacle avoidance motion planning and control of on-orbit redundant manipulators

    Hongjun Xing, Zeping Wang, Bin Lei, Yuyan Xie, Liang Ding, and Jinbao Chen. Hqp-based obstacle avoidance motion planning and control of on-orbit redundant manipulators. International Journal of Aeronautical and Space Sciences , Feb 2025. A Gradient of D-HLSP-E Constrained cont...

    2025

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.