arxiv: 2605.11043 · v1 · submitted 2026-05-11 · 🌌 astro-ph.IM · astro-ph.EP

Recognition: no theorem link

A practical guide to implementing zero-order-hold interplanetary trajectory legs

Dario Izzo , Harry Holt , Giacomo Acciarini , Laurent Beauregard , Yuri Shimane

Authors on Pith no claims yet

Pith reviewed 2026-05-13 01:08 UTC · model grok-4.3

classification 🌌 astro-ph.IM astro-ph.EP

keywords zero-order-hold transcriptiontrajectory optimizationspacecraft trajectoryinterplanetary missionsdynamical modelsbenchmark problemssoftmax encoding

0 comments

The pith

Zero-order-hold transcriptions become robust for interplanetary trajectory optimization across dynamical models without problem-specific tuning.

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

The paper investigates practical ways to implement zero-order-hold transcriptions in optimizing spacecraft trajectories. It identifies design principles including forward-backward shooting, a redundant throttle parameterization, and softmax time encoding that make these methods reliable in different settings like two-body and three-body problems. A sympathetic reader would care because current methods often require expert adjustments when changing dynamical models, leading to unreliable or time-consuming optimizations. The work provides evidence through a new benchmark suite of 28 problems.

Core claim

The authors establish that zero-order-hold transcriptions, using the forward-backward shooting construction denoted ZOH_alpha, a redundant four-dimensional throttle parameterization to remove control singularities on ballistic arcs, and a softmax time-grid encoding for differentiable segment durations, can be implemented robustly across a broad class of dynamical settings without problem-specific tuning.

What carries the argument

The ZOH_alpha forward-backward shooting construction, augmented by redundant four-dimensional throttle parameterization and softmax time-grid encoding.

If this is right

Trajectory optimization succeeds consistently in two-body Cartesian, modified equinoctial, circular restricted three-body, and solar sailing models.
Singularities in the control influence matrix are eliminated along ballistic arcs.
Segment durations can be optimized without ordering constraints while keeping the problem fully differentiable.
The TOPS benchmark provides 28 standardized problems for testing and extending trajectory optimization methods.

Where Pith is reading between the lines

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

Adopting these principles could allow non-experts to apply trajectory optimization more broadly in mission design.
The benchmark suite may facilitate direct comparisons between different optimization approaches in future studies.
Similar redundancy and encoding techniques might improve robustness in other types of optimal control problems beyond spacecraft trajectories.

Load-bearing premise

The identified design principles and techniques will generalize to new or unseen dynamical settings and problem instances without needing any problem-specific tuning or adjustments.

What would settle it

Demonstrating a new dynamical model or problem instance where applying the ZOH_alpha, redundant throttle, or softmax encoding without adjustments leads to optimization failure or singularity issues would falsify the robustness claim.

Figures

Figures reproduced from arXiv: 2605.11043 by Dario Izzo, Giacomo Acciarini, Harry Holt, Laurent Beauregard, Yuri Shimane.

**Figure 1.** Figure 1: Illustration of the ZOHα forward–backward transcription with N = 5, Nfwd = 3, Nbck = 2. Blue: forward arc propagated from x0; red: backward arc propagated from xf . Green and orange arrows indicate the constant ZOH thrust ui per segment. Triangular markers represent xfwd (blue, upward) and xbck (red, downward) at the shared junction time t3; the dashed gap is the mismatch mcα. and ∂xL ∂ti = ML,ih f(xi ,ui−… view at source ↗

**Figure 2.** Figure 2: Summary of all problems in the TOPS benchmark. a) [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗

**Figure 3.** Figure 3: The non-dominated fronts in the mean objective, ERT [s] metrics for all the TOPS problems in the [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗

**Figure 4.** Figure 4: top: Performance plots for all the TOPS problems in the [PITH_FULL_IMAGE:figures/full_fig_p012_4.png] view at source ↗

read the original abstract

We study the practical implementation of zero-order-hold (ZOH) transcriptions for spacecraft trajectory optimisation, identifying a set of design principles that render them robust across a broad class of dynamical settings without problem-specific tuning. The contributions are fourfold: (i) a thorough study of the forward--backward shooting construction, denoted $\mathrm{ZOH}_\alpha$; (ii) a redundant four-dimensional throttle parameterization that eliminates the singularity of the control influence matrix along ballistic arcs; (iii) a softmax time-grid encoding that avoids ordering constraints on segment durations while preserving full differentiability; and (iv) the TOPS benchmark (Trajectory Optimisation Problems in Space), a suite of 28 problems spanning four dynamical models, two-body Cartesian, modified equinoctial elements, circular restricted three-body, and solar sailing, designed to be extended over 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

The paper gives concrete implementation fixes for ZOH trajectory optimization plus a new benchmark set, but its no-tuning robustness claim rests only on the authors' own test problems.

read the letter

The main takeaway is that this paper collects a few targeted implementation choices that make zero-order-hold transcriptions less fragile in practice for interplanetary trajectories. The redundant four-dimensional throttle parameterization removes the singularity on ballistic arcs, the softmax time-grid encoding sidesteps ordering constraints while staying differentiable, and the forward-backward ZOH_alpha shooting gets a closer look. They also release the TOPS suite of 28 problems across four dynamical models as a starting point for comparisons. These are the parts that feel genuinely new and useful for anyone who has to code these things up. The work does well at staying practical. It focuses on the numerical and structural details that cause trouble in real optimizers rather than abstract theory, and the benchmark is set up to be extended, which is a reasonable contribution on its own. The authors appear to have thought through the differentiability and conditioning issues that come up in actual runs. The soft spot is the central claim about robustness without problem-specific tuning. The evidence comes from success on the 28 TOPS instances, but those cases were selected and likely iterated on while the principles were being developed. That makes it hard to know whether the same fixed implementation will transfer cleanly to new models, different scales, or unseen constraints. No hold-out set or reported failures on external problems are described, so the generalization argument stays internal to the authors' testbed. This paper is for people who build or maintain trajectory optimization code for mission design. A reader who needs to implement ZOH legs or wants a common set of test cases will get direct value from the details and the benchmark. It is not a theoretical breakthrough, but the implementation focus is solid enough that it deserves a serious referee. I would send it out for peer review rather than desk reject.

Referee Report

1 major / 2 minor

Summary. The manuscript studies practical implementation of zero-order-hold (ZOH) transcriptions for spacecraft trajectory optimization. It identifies four design principles claimed to yield robust performance across dynamical settings without problem-specific tuning: (i) the ZOH_α forward-backward shooting construction, (ii) a redundant 4D throttle parameterization to remove singularities along ballistic arcs, (iii) a softmax time-grid encoding that preserves differentiability without ordering constraints, and (iv) the TOPS benchmark suite of 28 problems spanning two-body Cartesian, modified equinoctial, CR3BP, and solar-sailing models.

Significance. If the robustness claims hold, the work supplies a concrete, reusable implementation template and an extensible benchmark (TOPS) that could reduce ad-hoc tuning in interplanetary trajectory design. The explicit construction of ZOH_α, the 4D throttle map, and the differentiable softmax grid are useful engineering contributions; the benchmark's stated extensibility is a strength that future authors can build upon.

major comments (1)

[§5] §5 (TOPS benchmark results): The central claim that the four principles produce robust performance 'without problem-specific tuning' on a broad class of problems rests on performance across the 28 TOPS instances. Because these instances were constructed and possibly refined during principle development, success on them does not demonstrate that the identical fixed implementation will succeed on out-of-distribution problems (new models, scales, or constraints) without adjustment. No hold-out set, external benchmark, or reported failure modes on unseen instances are described.

minor comments (2)

[Abstract / Introduction] The abstract lists four contributions but does not indicate where each is formally stated or proved; a short numbered list or theorem-style summary at the end of the introduction would improve traceability.
[§3] Notation for the ZOH_α construction and the 4D throttle vector should be introduced once with a single table of symbols rather than redefined inline in multiple sections.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their careful reading and constructive critique of our manuscript. We address the major comment below and outline the revisions we will make.

read point-by-point responses

Referee: [§5] §5 (TOPS benchmark results): The central claim that the four principles produce robust performance 'without problem-specific tuning' on a broad class of problems rests on performance across the 28 TOPS instances. Because these instances were constructed and possibly refined during principle development, success on them does not demonstrate that the identical fixed implementation will succeed on out-of-distribution problems (new models, scales, or constraints) without adjustment. No hold-out set, external benchmark, or reported failure modes on unseen instances are described.

Authors: We agree that the TOPS instances were assembled iteratively during the development of the four design principles and therefore do not constitute an independent hold-out set. The principles themselves (ZOH_α forward-backward shooting, redundant 4D throttle map, softmax time-grid encoding, and the overall transcription structure) were motivated by analysis of the underlying mathematical properties of zero-order-hold discretizations—specifically, consistency of the shooting map, removal of control singularities on ballistic arcs, and preservation of differentiability—rather than by empirical fitting to particular problems. Nevertheless, the referee correctly identifies that this development process weakens the strength of the generalization statement as currently phrased. In the revised manuscript we will (i) add an explicit description of the iterative construction of TOPS in §5, (ii) qualify the robustness claim to state that the fixed implementation succeeded without tuning on the 28 problems spanning the four dynamical models, and (iii) include a short discussion of observed failure modes during benchmark development together with guidance on how the same fixed encoding can be applied to new problems. These changes will make the presentation more precise while retaining the benchmark as an extensible community resource. revision: partial

Circularity Check

0 steps flagged

No circularity: practical implementation guide with new benchmark

full rationale

The paper presents four concrete implementation contributions (ZOH_alpha shooting construction, 4D throttle parameterization, softmax time-grid encoding, and the TOPS benchmark suite) and demonstrates their performance on the 28 TOPS instances. These are introduced as design choices and a new test suite rather than derived quantities. No equations reduce a claimed result to its own inputs by construction, no parameters are fitted to a subset and then relabeled as predictions, and no load-bearing self-citations or uniqueness theorems imported from prior author work are used to justify the central claims. The robustness statements rest on empirical results across the introduced benchmark, which is an independent (if author-constructed) validation set rather than a self-referential loop.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The work is an implementation guide relying on standard optimal control and numerical optimization methods rather than introducing new fitted parameters, axioms, or postulated entities.

axioms (1)

standard math Standard assumptions of optimal control theory and differentiability of the dynamics and cost functions
ZOH transcriptions and gradient-based optimization presuppose these properties for the forward-backward shooting construction to be valid.

pith-pipeline@v0.9.0 · 5450 in / 1199 out tokens · 43533 ms · 2026-05-13T01:08:24.762772+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

30 extracted references · 30 canonical work pages

[1]

Survey of numerical methods for tra- jectory optimization

Betts, J. T. “Survey of numerical methods for tra- jectory optimization.” Journal of guidance, control, and dynamics, Vol. 21, No. 2, pp. 193–207, 1998

work page 1998
[2]

Direct trajec- tory optimization using nonlinear programming and collocation

Hargraves, C. R. and Paris, S. W. “Direct trajec- tory optimization using nonlinear programming and collocation.” Journal of guidance, control, and dy- namics, Vol. 10, No. 4, pp. 338–342, 1987

work page 1987
[3]

Survey of Direct Transcription for Low-Thrust Space Trajectory Op- timization with Applications

Topputo, F. and Zhang, C. “Survey of Direct Transcription for Low-Thrust Space Trajectory Op- timization with Applications.” Abstract and Ap- plied Analysis, Vol. 2014, p. 851720, 2014. doi: 10.1155/2014/851720

work page doi:10.1155/2014/851720 2014
[4]

Direct Trajec- tory Optimization by a Chebyshev Pseudospectral Method

Fahroo, F. and Ross, I. M. “Direct Trajec- tory Optimization by a Chebyshev Pseudospectral Method.” Journal of Guidance, Control, and Dy- namics, Vol. 25, No. 1, pp. 160–166, 2002. doi: 10.2514/2.4862

work page doi:10.2514/2.4862 2002
[5]

GPOPS-II: A MATLAB Software for Solving Multiple-Phase Op- timal Control Problems Using hp-Adaptive Gaus- sian Quadrature Collocation Methods and Sparse Nonlinear Programming

Patterson, M. A. and Rao, A. V. “GPOPS-II: A MATLAB Software for Solving Multiple-Phase Op- timal Control Problems Using hp-Adaptive Gaus- sian Quadrature Collocation Methods and Sparse Nonlinear Programming.” ACM Transactions on Mathematical Software, Vol. 41, No. 1, p. 1, 2014. doi:10.1145/2558904

work page doi:10.1145/2558904 2014
[6]

S., Boltyanskii, V

Pontryagin, L. S., Boltyanskii, V. G., Gamkrelidze, R. V., and Mishchenko, E. F. The Mathematical Theory of Optimal Processes. Wiley-Interscience, New York, 1962. Translated from the Russian by K. N. Trirogoff

work page 1962
[7]

Prelimi- nary Design of Low-Thrust Interplanetary Mis- sions

Sims, J. A. and Flanagan, S. N. “Prelimi- nary Design of Low-Thrust Interplanetary Mis- sions.” “AAS/AIAA Astrodynamics Specialist Conference,” 1997. NASA Technical Report N20000057422

work page 1997
[8]

Implementa- tion of a Low-Thrust Trajectory Optimization Algo- rithm for Preliminary Design

Sims, J. A., Finlayson, P. A., Rinderle, E. A., Vav- rina, M. A., and Kowalkowski, T. D. “Implementa- tion of a Low-Thrust Trajectory Optimization Algo- rithm for Preliminary Design.” 2006. Paper AIAA 2006-6746

work page 2006
[9]

PyGMO and PyKEP: Open Source Tools for Massively Parallel Optimization in Astrodynam- ics (the Case of Interplanetary Trajectory Opti- mization)

Izzo, D. “PyGMO and PyKEP: Open Source Tools for Massively Parallel Optimization in Astrodynam- ics (the Case of Interplanetary Trajectory Opti- mization).” “Proceedings of the Fifth International Conference on Astrodynamics Tools and Techniques (ICATT),” ESA, 2012

work page 2012
[10]

Towards a High Fidelity Direct Transcription Method for Optimisation of Low-Thrust Trajectories

Yam, C. H., Izzo, D., and Biscani, F. “Towards a High Fidelity Direct Transcription Method for Optimisation of Low-Thrust Trajectories.” arXiv preprint, 2010

work page 2010
[11]

Low- Thrust Trajectory Design as a Constrained Global Optimization Problem

Yam, C. H., Di Lorenzo, D., and Izzo, D. “Low- Thrust Trajectory Design as a Constrained Global Optimization Problem.” Proceedings of the Insti- tution of Mechanical Engineers, Part G: Journal of Aerospace Engineering, Vol. 225, No. 11, pp. 1243– 1251, 2011. doi:10.1177/0954410011401686

work page doi:10.1177/0954410011401686 2011
[12]

Convex optimization for trajectory generation: A tutorial on generating dynamically feasible trajectories reliably and efficiently

Malyuta, D., Reynolds, T. P., Szmuk, M., Lew, T., Bonalli, R., Pavone, M., and A¸ cıkme¸ se, B. “Convex optimization for trajectory generation: A tutorial on generating dynamically feasible trajectories reliably and efficiently.” IEEE Control Systems Magazine, Vol. 42, No. 5, pp. 40–113, 2022

work page 2022
[13]

Convex program- ming approach to powered descent guidance for mars landing

A¸ cıkme¸ se, B. and Ploen, S. R. “Convex program- ming approach to powered descent guidance for mars landing.” Journal of Guidance, Control, and Dynamics, Vol. 30, No. 5, pp. 1353–1366, 2007

work page 2007
[14]

Lossless convexification of nonconvex control bound and pointing constraints of the soft land- ing optimal control problem

A¸ cıkme¸ se, B., Carson, J. M., and Blackmore, L. “Lossless convexification of nonconvex control bound and pointing constraints of the soft land- ing optimal control problem.” IEEE transactions on control systems technology, Vol. 21, No. 6, pp. 2104–2113, 2013

work page 2013
[15]

Successive convexification of non-convex optimal control prob- lems and its convergence properties

Mao, Y., Szmuk, M., and A¸ cıkme¸ se, B. “Successive convexification of non-convex optimal control prob- lems and its convergence properties.” “2016 IEEE 55th Conference on Decision and Control (CDC),” pp. 3636–3641. IEEE, 2016

work page 2016
[16]

Successive convexification with feasi- bility guarantee via augmented lagrangian for non- convex optimal control problems

Oguri, K. “Successive convexification with feasi- bility guarantee via augmented lagrangian for non- convex optimal control problems.” “2023 62nd IEEE Conference on Decision and Control (CDC),” pp. 3296–3302. IEEE, 2023

work page 2023
[17]

Practical tech- niques for low-thrust trajectory optimization with homotopic approach

Jiang, F., Baoyin, H., and Li, J. “Practical tech- niques for low-thrust trajectory optimization with homotopic approach.” Journal of guidance, control, and dynamics, Vol. 35, No. 1, pp. 245–258, 2012

work page 2012
[18]

Regularized direct method for low–thrust trajectory optimization: Minimum–fuel transfer between cislunar periodic orbits

Oshima, K. “Regularized direct method for low–thrust trajectory optimization: Minimum–fuel transfer between cislunar periodic orbits.” Advances in Space Research, Vol. 72, No. 6, pp. 2051–2063, 2023

work page 2051
[19]

An Auto- mated Solution of the Low-Thrust Interplanetary Trajectory Problem

Englander, J. A. and Conway, B. A. “An Auto- mated Solution of the Low-Thrust Interplanetary Trajectory Problem.” Journal of Guidance, Con- trol, and Dynamics, Vol. 40, No. 1, pp. 15–27, 2017. doi:10.2514/1.G002124

work page doi:10.2514/1.g002124 2017
[20]

Per- formance Assessment of Convex Low-Thrust Trajec- tory Optimization Methods

Hofmann, C., Morelli, A. C., and Topputo, F. “Per- formance Assessment of Convex Low-Thrust Trajec- tory Optimization Methods.” Journal of Spacecraft and Rockets, Vol. 60, No. 1, pp. 299–314, 2023. doi: 10.2514/1.A35461

work page doi:10.2514/1.a35461 2023
[21]

A comparison of SEP and NEP for a main belt asteroid sample return mission

Dachwald, B., Seboldt, W., Loeb, H. W., and Schartner, K.-H. “A comparison of SEP and NEP for a main belt asteroid sample return mission.” “Proceedings of the 7th International Symposium on Launcher Technologies, Barcelona, Spain,” 2007

work page 2007
[22]

Global optimization approaches for optimal trajectory planning

Cassioli, A., Izzo, D., Di Lorenzo, D., Locatelli, M., and Schoen, F. “Global optimization approaches for optimal trajectory planning.” “Modeling and optimization in space engineering,” pp. 111–140. Springer, 2012

work page 2012
[23]

A High-Precision, Differentiable Code for Solar System Ephemerides

Cassese, B., Rice, M., and Lu, T. “A High-Precision, Differentiable Code for Solar System Ephemerides.” The Planetary Science Journal, Vol. 6, 2025. doi: 10.3847/PSJ/adaaf9. ArXiv:2509.19549

work page doi:10.3847/psj/adaaf9 2025
[24]

En- hanced smoothing technique for indirect optimiza- tion of minimum-fuel low-thrust trajectories

Taheri, E., Kolmanovsky, I., and Atkins, E. “En- hanced smoothing technique for indirect optimiza- tion of minimum-fuel low-thrust trajectories.” Jour- nal of Guidance, Control, and Dynamics, Vol. 39, No. 11, pp. 2500–2511, 2016

work page 2016
[25]

Revisiting high-order Tay- lor methods for astrodynamics and celestial mechan- ics

Biscani, F. and Izzo, D. “Revisiting high-order Tay- lor methods for astrodynamics and celestial mechan- ics.” Monthly Notices of the Royal Astronomical Society, Vol. 504, No. 2, pp. 2614–2628, 2021

work page 2021
[26]

On the implementa- tion of an interior-point filter line-search algorithm for large-scale nonlinear programming

W¨ achter, A. and Biegler, L. T. “On the implementa- tion of an interior-point filter line-search algorithm for large-scale nonlinear programming.” Mathemat- ical programming, Vol. 106, No. 1, pp. 25–57, 2006

work page 2006
[27]

SNOPT: An SQP algorithm for large-scale con- strained optimization

Gill, P. E., Murray, W., and Saunders, M. A. “SNOPT: An SQP algorithm for large-scale con- strained optimization.” SIAM review, Vol. 47, No. 1, pp. 99–131, 2005

work page 2005
[28]

Analyzing the BBOB Results by Means of Benchmarking Concepts

Mersmann, O., Preuss, M., Trautmann, H., Bischl, B., and Weihs, C. “Analyzing the BBOB Results by Means of Benchmarking Concepts.” Evolutionary Computation, Vol. 23, No. 1, pp. 161–185, 2015. doi:10.1162/EVCO a 00134

work page doi:10.1162/evco 2015
[29]

Benchmarking Optimization Software with Performance Profiles

Dolan, E. D. and Mor´ e, J. J. “Benchmarking Optimization Software with Performance Profiles.” Mathematical Programming, Vol. 91, No. 2, pp. 201–213, 2002. doi:10.1007/s101070100263

work page doi:10.1007/s101070100263 2002
[30]

A survey of numerical methods for op- timal control

Rao, A. V. “A survey of numerical methods for op- timal control.” Advances in the Astronautical Sci- ences, Vol. 135, No. 1, pp. 497–528, 2009. Fig. 2: Summary of all problems in the TOPS benchmark. a)twobodythe basic Cartesian spacecraft dynamics, b) meesame, but using Mean Equinoctial Elements, c)cr3bpthe circular restricted three body problem and d)sst...

work page 2009