pith. sign in

arxiv: 2606.28207 · v1 · pith:YZT4HF5Nnew · submitted 2026-06-26 · 🧮 math.OC

Three-Body Earth-Moon Transfers with Different Departure/Arrival Orbital Altitudes: New Phenomenon and Diffusion Model-Augmented Construction

Shuyue Fu , Wenxuan Zhang , Di Wu , Shengping Gong , Peng Shi This is my paper

Pith reviewed 2026-06-29 02:51 UTC · model grok-4.3

classification 🧮 math.OC
keywords Earth-Moon transferdiffusion modelgrid searchthree-body problemtime of flightinitial guessorbital altitude
0
0 comments X

The pith

Diffusion model trained on time-of-flight discontinuities augments grid search for Earth-Moon transfers at varying altitudes.

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

This paper constructs bi-impulsive Earth-Moon transfers in the planar circular restricted three-body problem. It identifies a discontinuous pattern in the distribution of time of flight as a function of departure phase angle. The authors train a diffusion model on this pattern to generate initial guesses. The model augments the traditional grid search, enabling construction of transfers for different Earth parking orbit and Moon target orbit altitudes. The resulting method raises convergence rates and lowers wall-clock time while producing transfers with characteristics comparable to those from repeated standard grid searches.

Core claim

The discontinuous behavior of the time-of-flight distribution with respect to departure phase angle can be used to train a diffusion model that generates high-quality initial guesses, allowing efficient construction of transfers with different departure and arrival orbital altitudes via an augmented grid search method.

What carries the argument

A diffusion model trained on the observed discontinuous time-of-flight versus departure phase angle distribution, which generates initial guesses to augment the grid search process.

If this is right

  • Transfers at new altitude pairs can be constructed without repeating the full grid search and correction process from scratch.
  • The search convergence rate increases by 47.34-56.25 percent relative to traditional grid search.
  • Wall-clock time is reduced by 39.39-40.52 percent relative to traditional grid search.
  • The resulting transfers maintain transfer characteristics comparable to those obtained by repeated standard searches.

Where Pith is reading between the lines

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

  • The same discontinuity-to-model pipeline could be applied to other three-body or multi-body trajectory families that exhibit phase-angle sensitivities.
  • Real-time replanning for cislunar missions might become feasible if the trained model generalizes across wide altitude ranges without retraining.
  • Hybrid methods that combine diffusion-based guess generation with direct optimization could extend the approach to continuous-thrust or low-thrust Earth-Moon transfers.

Load-bearing premise

The discontinuous time-of-flight behavior seen at one altitude set is representative enough that a model trained on it works well for transfers at substantially different altitudes.

What would settle it

Running the diffusion model-augmented search on a new set of departure and arrival altitudes and finding no improvement in convergence rate or computation time over standard grid search would falsify the utility of the approach.

Figures

Figures reproduced from arXiv: 2606.28207 by Di Wu, Peng Shi, Shengping Gong, Shuyue Fu, Wenxuan Zhang.

Figure 1
Figure 1. Figure 1: The (TOF, Δ𝑣) map of obtained bi-impulsive Earth-Moon transfers in the Earth-Moon PCR3BP. (a) Group I; (b) Group II; (c) Group III. For the obtained solutions, the ( TOF, 𝛼𝑖 ) map and the ( 𝛼𝑖 , 𝛽𝑖 ) map for three groups are presented in Figs. 2-3. The construction parameters for these four groups exhibit similar distributions. Differing from the ( 𝛼𝑖 , 𝛽𝑖 ) map, the ( TOF, 𝛼𝑖 ) map reveals a rather pronou… view at source ↗
Figure 2
Figure 2. Figure 2: The ( TOF, 𝛼𝑖 ) map of obtained bi-impulsive Earth-Moon transfers in the Earth-Moon PCR3BP. (a) Group I; (b) Group II; (c) Group III. Shuyue Fu et al.: Preprint submitted to Elsevier Page 6 of 24 [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The ( 𝛼𝑖 , 𝛽𝑖 ) map of obtained bi-impulsive Earth-Moon transfers in the Earth-Moon PCR3BP. (a) Group I; (b) Group II; (c) Group III. The aforementioned phenomenon exists in all of three groups, implying its robustness with respect to the orbital altitude settings. Also, when adopting a smaller step-size of construction parameters, a similar phenomenon has been observed for Group I [34], implying its robus… view at source ↗
Figure 4
Figure 4. Figure 4: Solution space for different groups. (a) [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: The combined ( TOF, 𝛼𝑖 ) map for Group I. Once the dataset (i.e., the transformed ( TOF, 𝛼𝑖 ) map shown in [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Schematic of the forward diffusion process and reverse denoising process. [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Schematic of the structure of the neural network. [PITH_FULL_IMAGE:figures/full_fig_p011_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: The loss time history for the aforementioned hyperparameter combinations. [PITH_FULL_IMAGE:figures/full_fig_p014_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: The loss time history of retraining. Shuyue Fu et al.: Preprint submitted to Elsevier Page 15 of 24 [PITH_FULL_IMAGE:figures/full_fig_p015_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: The generated samples. Then, with the generated samples, the proposed diffusion model-augmented grid search method is applied to the construction of bi-impulsive Earth-Moon transfers. Notably, the application of the generated samples is not only limited to constructing Earth-Moon transfers with ℎ𝑖 = 167 km and ℎ𝑓 = 100 km, but also with Groups II and III because of the similar ( TOF, 𝛼𝑖 ) distributions sh… view at source ↗
Figure 11
Figure 11. Figure 11: The (TOF, Δ𝑣) map of bi-impulsive Earth-Moon transfers in the Earth-Moon PCR3BP obtained from the diffusion model-augmented grid search method. (a) Group I; (b) Group II; (c) Group III. We extract the Pareto front of the (TOF, Δ𝑣) map shown in [PITH_FULL_IMAGE:figures/full_fig_p016_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Pareto front of the (TOF, Δ𝑣) map (Group I) [PITH_FULL_IMAGE:figures/full_fig_p017_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Pareto front of the (TOF, Δ𝑣) map (Group II) [PITH_FULL_IMAGE:figures/full_fig_p017_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Pareto front of the (TOF, Δ𝑣) map (Group III). Shuyue Fu et al.: Preprint submitted to Elsevier Page 17 of 24 [PITH_FULL_IMAGE:figures/full_fig_p017_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Selected trajectory sample (Group I). Shuyue Fu et al.: Preprint submitted to Elsevier Page 18 of 24 [PITH_FULL_IMAGE:figures/full_fig_p018_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: Selected trajectory sample (Group II). Shuyue Fu et al.: Preprint submitted to Elsevier Page 19 of 24 [PITH_FULL_IMAGE:figures/full_fig_p019_16.png] view at source ↗
Figure 17
Figure 17. Figure 17: Selected trajectory sample (Group III). Subsequently, the comparison between results obtained from the proposed diffusion model-augmented grid search method and the traditional grid search method (mentioned in Section 2.3) is performed to reveal the advantages of the developed method. 4.3. Comparison with the Traditional Grid Search Method The comparison with the traditional grid search method can be divi… view at source ↗
Figure 18
Figure 18. Figure 18: Pareto fronts obtained from the two methods. (a) Group I; (b) Group II; (c) Group III. [PITH_FULL_IMAGE:figures/full_fig_p021_18.png] view at source ↗
read the original abstract

Construction of Earth-Moon transfers is the basis of missions to explore the Moon and cislunar space. The traditional grid search method suffers from a relatively low convergence rate and computational efficiency, mainly focusing on the distribution of transfer characteristic parameters. Moreover, when constructing transfers with different departure/arrival orbital altitudes, the process of grid search and trajectory correction should be repeated with a low convergence rate and computational efficiency. To address these limitations of the traditional grid search method, this paper is devoted to exploring an effective way to augment the grid search method. Bi-impulsive Earth-Moon transfers from a circular Earth parking orbit to a circular Moon target orbit in the Earth-Moon planar circular restricted three-body problem are considered in this paper. Firstly, the transfers are constructed, and the corresponding solution space is explored in terms of construction parameters, including departure phase angle at the Earth parking orbit, initial-to-circular velocity ratio, and time of flight. An interesting phenomenon about the discontinuous behavior of the time-of-flight distribution with respect to departure phase angle is identified. This phenomenon is further used to train a diffusion model, which aims to augment the traditional grid search method and generate high-quality initial guesses for transfers with different departure/arrival orbital altitudes. The construction results of the proposed method are presented and analyzed. The proposed diffusion model-augmented grid search method improves the convergence rate by 47.34-56.25% and saves the wall-clock time by 39.39-40.52% over the traditional grid search method relatively, while ensuring comparable transfer characteristics.

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 / 1 minor

Summary. The paper considers bi-impulsive Earth-Moon transfers in the planar circular restricted three-body problem from circular Earth parking orbits to circular Moon target orbits. It identifies a discontinuous dependence of time-of-flight on departure phase angle at one specific pair of orbital altitudes, trains a diffusion model on this behavior, and uses the model to supply initial guesses that augment a traditional grid-search procedure when the departure and arrival altitudes are changed. The abstract reports that the augmented method raises convergence rate by 47.34–56.25 % and reduces wall-clock time by 39.39–40.52 % relative to unaugmented grid search while preserving comparable transfer characteristics.

Significance. If the diffusion model generalizes the observed discontinuity across altitude pairs, the method would offer a concrete, reusable way to accelerate grid-based construction of three-body transfers. The reported speed-ups are empirical and therefore directly relevant to mission-design workflows that repeatedly solve similar problems at varying radii.

major comments (2)
  1. [Abstract] Abstract: the headline performance figures (47.34–56.25 % convergence improvement, 39.39–40.52 % wall-time reduction) are presented without error bars, without stating how many distinct altitude pairs were tested, and without indicating whether the diffusion model was trained or tuned on any of the reported test cases; these omissions make the central empirical claim impossible to assess for statistical reliability.
  2. [Abstract] Abstract: the diffusion model is trained exclusively on the discontinuous TOF-versus-phase-angle behavior observed at one Earth/Moon altitude pair and then applied to “different departure/arrival orbital altitudes,” yet no cross-validation, ablation, or structural analysis is supplied to show that the discontinuity persists or retains comparable geometry when the radii change; because this generalization is the load-bearing assumption for the claimed speed-ups, its absence undermines the transferability of the reported gains.
minor comments (1)
  1. The phrase “comparable transfer characteristics” is used without defining the quantitative metrics (e.g., Δv, TOF, or Jacobi constant) or tolerance thresholds employed for the comparison.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments on our manuscript. We address each major comment point by point below and will revise the manuscript accordingly to improve clarity and strengthen the empirical claims.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the headline performance figures (47.34–56.25 % convergence improvement, 39.39–40.52 % wall-time reduction) are presented without error bars, without stating how many distinct altitude pairs were tested, and without indicating whether the diffusion model was trained or tuned on any of the reported test cases; these omissions make the central empirical claim impossible to assess for statistical reliability.

    Authors: We acknowledge the need for greater transparency in the abstract. The reported performance ranges are aggregated from experiments across multiple distinct altitude pairs (distinct from the single training pair), with the diffusion model trained exclusively on the identified discontinuous case at one altitude pair and applied without further tuning to the test cases. To address this, we will revise the abstract to explicitly state the number of altitude pairs tested, clarify the training/test separation, and include error bars or standard deviations from repeated runs where applicable. revision: yes

  2. Referee: [Abstract] Abstract: the diffusion model is trained exclusively on the discontinuous TOF-versus-phase-angle behavior observed at one Earth/Moon altitude pair and then applied to “different departure/arrival orbital altitudes,” yet no cross-validation, ablation, or structural analysis is supplied to show that the discontinuity persists or retains comparable geometry when the radii change; because this generalization is the load-bearing assumption for the claimed speed-ups, its absence undermines the transferability of the reported gains.

    Authors: The manuscript states that the model is trained on the discontinuity at one specific altitude pair and then used for different altitudes, with the results demonstrating successful augmentation and the reported speed-ups. We agree that additional evidence of generalization would strengthen the work. We will add a dedicated subsection providing structural analysis of the discontinuity across altitude variations, along with cross-validation results on held-out altitude pairs, to explicitly support transferability. revision: yes

Circularity Check

0 steps flagged

No significant circularity; empirical gains are externally measured

full rationale

The paper identifies a discontinuous TOF-vs-phase-angle pattern at one specific Earth/Moon altitude pair, trains a diffusion model on that data, and then reports measured convergence-rate and wall-time improvements when the model supplies initial guesses for transfers at different altitude pairs. These performance numbers are direct empirical comparisons against a baseline grid-search implementation; they are not algebraically or definitionally forced by any fitted parameter inside the diffusion model, nor do they reduce to a self-citation chain or an ansatz smuggled from prior work by the same authors. The derivation chain therefore remains self-contained against external computational benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The paper relies on the standard planar circular restricted three-body problem (CR3BP) equations and on the assumption that the discontinuous TOF pattern is a stable feature of the dynamics; no additional free parameters or invented entities are stated in the abstract.

axioms (1)
  • domain assumption The planar circular restricted three-body problem accurately models the dominant dynamics for bi-impulsive Earth-Moon transfers.
    Invoked implicitly when the authors state they work inside the Earth-Moon planar CR3BP.

pith-pipeline@v0.9.1-grok · 5832 in / 1352 out tokens · 56970 ms · 2026-06-29T02:51:42.252357+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

36 extracted references · 30 canonical work pages · 2 internal anchors

  1. [1]

    Topputo, On optimal two-impulse earth–moon transfers in a four-body model, Celest

    F. Topputo, On optimal two-impulse earth–moon transfers in a four-body model, Celest. Mech. Dyn. Astron. 117 (2013) 279–313.doi: 10.1007/s10569-013-9513-8

    work page doi:10.1007/s10569-013-9513-8 2013

  2. [2]

    doi:10.1016/j.actaastro.2017.02.002

    Y.Qi,S.Xu,Optimalearth-moontransfersusinglunargravityassistintherestrictedfour-bodyproblem,ActaAstronaut.134(2017)106–120. doi:10.1016/j.actaastro.2017.02.002

    work page doi:10.1016/j.actaastro.2017.02.002 2017

  3. [3]

    Oshima, F

    K. Oshima, F. Topputo, T. Yanao, Low-energy transfers to the moon with long transfer time, Celest. Mech. Dyn. Astron. 131 (2019) 1–19. doi:10.1007/s10569-019-9883-7

    work page doi:10.1007/s10569-019-9883-7 2019

  4. [4]

    Demeyer, P

    J. Demeyer, P. Gurfil, Transfer to distant retrograde orbits using manifold theory, J. Guid. Control Dyn. 30 (5) (2007) 1261–1267.doi: 10.2514/1.24960

    work page doi:10.2514/1.24960 2007

  5. [5]

    Y.Yin,M.Wang,Y.Shi,H.Zhang,Midcoursecorrectionofearth-moondistantretrogradeorbittransfertrajectoriesbasedonhigh-orderstate transition tensors, Astrodynamics 7 (3) (2023) 335–349.doi:10.1007/s42064-023-0162-8

    work page doi:10.1007/s42064-023-0162-8 2023

  6. [6]

    M. Wang, C. Zhang, H. Zhang, Mechanism analysis of the dro low-energy transfer problem: An energy perspective, Astrodynamics 9 (2) (2025) 165–193.doi:10.1007/s42064-024-0215-7

    work page doi:10.1007/s42064-024-0215-7 2025

  7. [7]

    Mingotti, P

    G. Mingotti, P. Gurfil, Mixed low-thrust invariant-manifold transfers to distant prograde orbits around mars, J. Guid. Control Dyn. 33 (6) (2010) 1753–1764.doi:10.2514/1.49810

    work page doi:10.2514/1.49810 2010

  8. [8]

    79 (2012) 20–32.doi:10.1016/j.actaastro.2012.04.022

    G.Mingotti,F.Topputo,F.Bernelli-Zazzera,Transferstodistantperiodicorbitsaroundthemoonviatheirinvariantmanifolds,ActaAstronaut. 79 (2012) 20–32.doi:10.1016/j.actaastro.2012.04.022

    work page doi:10.1016/j.actaastro.2012.04.022 2012

  9. [9]

    J. S. Parker, R. L. Anderson, Low-energy lunar trajectory design, Vol. 12, John Wiley & Sons, 2014

    2014

  10. [10]

    C. Du, O. L. Starinova, Y. Liu, Transfer between the planar lyapunov orbits around the earth-moon l2 point using low-thrust engine, Acta Astronaut. 201 (2022) 513–525.doi:10.1016/j.actaastro.2022.09.056

    work page doi:10.1016/j.actaastro.2022.09.056 2022

  11. [11]

    Torralbo, I

    K. Oshima, A graph-based framework of low-energy transfer design, Acta Astronaut. 229 (2025) 644–661.doi:10.1016/j.actaastro. 2025.01.050

    work page doi:10.1016/j.actaastro 2025

  12. [12]

    Zhang, X

    Z. Zhang, X. Hou, Transfer orbits to the earth–moon triangular libration points, Adv. Space Res. 55 (12) (2015) 2899–2913.doi: 10.1016/j.asr.2015.03.008

    work page doi:10.1016/j.asr.2015.03.008 2015

  13. [13]

    L. R. Capdevila, K. C. Howell, A transfer network linking earth, moon, and the triangular libration point regions in the earth-moon system, Adv. Space Res. 62 (7) (2018) 1826–1852.doi:10.1016/j.asr.2018.06.045

    work page doi:10.1016/j.asr.2018.06.045 2018

  14. [14]

    Y. Yin, Y. Shi, H. Zhang, A graph-based framework for robust low-energy transfer design in multi-body system, Acta Astronaut. 248 (2026) 449–467.doi:10.1016/j.actaastro.2026.06.018

    work page doi:10.1016/j.actaastro.2026.06.018 2026

  15. [15]

    R. H. Battin, An introduction to the mathematics and methods of astrodynamics, AIAA, 1999

    1999

  16. [16]

    Pernicka, D

    H. Pernicka, D. Scarberry, S. Marsh, T. Sweetser, A search for low delta-v earth-to-moon trajectories, in: Astrodynamics Conference, AIAA- 94-3772-CP, AIAA, Scottsdale, 1994, pp. 530–537.doi:10.2514/6.1994-3772

    work page doi:10.2514/6.1994-3772 1994

  17. [17]

    Q. Guo, H. Lei, Families of earth–moon trajectories with applications to transfers towards sun–earth libration point orbits, Astrophys. Space Sci. 364 (3) (2019) 43.doi:10.1007/s10509-019-3532-1

    work page doi:10.1007/s10509-019-3532-1 2019

  18. [18]

    H. Jing, Z. Du, H. Li, H. Wang, J. He, C. Huo, Study on the reachable domain of earth–moon transfer orbits with terminal constraints in the cr3bp, Acta Astronaut. (2026).doi:10.1016/j.actaastro.2026.03.046

    work page doi:10.1016/j.actaastro.2026.03.046 2026

  19. [19]

    E. A. Belbruno, J. K. Miller, Sun-perturbed earth-to-moon transfers with ballistic capture, J. Guid. Control Dyn. 16 (4) (1993) 770–775. doi:10.2514/3.21079

    work page doi:10.2514/3.21079 1993

  20. [20]

    C. Wang, W. Liu, Y. Gao, Low-energy earth–moon transfer autonomous guidance considering high-fidelity orbital dynamics, Astrodynamics 8 (4) (2024) 689–701.doi:10.1007/s42064-024-0211-y

    work page doi:10.1007/s42064-024-0211-y 2024

  21. [21]

    C.T.Campana,F.Topputo,Ephemerisrefinementoflow-energyearth-moontransfersviaanastrodynamicmodelchain,ActaAstronaut.232 (2025) 271–282.doi:10.1016/j.actaastro.2025.03.010

    work page doi:10.1016/j.actaastro.2025.03.010 2025

  22. [22]

    W. S. Koon, M. W. Lo, J. E. Marsden, S. D. Ross, Low energy transfer to the moon, Celest. Mech. Dyn. Astron. 81 (1-2) (2001) 63–73. doi:10.1023/A:1013359120468

    work page doi:10.1023/a:1013359120468 2001

  23. [23]

    S. Fu, D. Wu, P. Shi, S. Gong, Four-body transit-orbit classification combining lagrangian coherent structures with data mining, J. Guid. Control Dyn. (2025) 1–20doi:10.2514/1.G008681. Shuyue Fu et al.:Preprint submitted to ElsevierPage 23 of 24 Diffusion Model-Augmented Construction of Three-Body Earth-Moon T ransfers

    work page doi:10.2514/1.g008681 2025

  24. [24]

    Hiraiwa, M

    N. Hiraiwa, M. Bando, Y. Sato, S. Hokamoto, Design of low-energy transfers in cislunar space using sequences of lobe dynamics, Acta Astronaut. (2026).doi:10.1016/j.actaastro.2026.05.032

    work page doi:10.1016/j.actaastro.2026.05.032 2026

  25. [25]

    S. Fu, Z. Xu, D. Wu, S. Gong, Deep neural network-based high-precision identification of weak stability boundary structures, IEEE Trans. Aerosp. Electron. Syst. (2026).doi:10.1109/TAES.2026.3668754

    work page doi:10.1109/taes.2026.3668754 2026

  26. [26]

    Oshima, F

    K. Oshima, F. Topputo, S. Campagnola, T. Yanao, Analysis of medium-energy transfers to the moon, Celest. Mech. Dyn. Astron. 127 (3) (2017) 285–300.doi:10.1007/s10569-016-9727-7

    work page doi:10.1007/s10569-016-9727-7 2017

  27. [27]

    Dutt, A review of low-energy transfers, Astrophys

    P. Dutt, A review of low-energy transfers, Astrophys. Space Sci. 363 (12) (2018) 253.doi:10.1007/s10509-018-3461-4

    work page doi:10.1007/s10509-018-3461-4 2018

  28. [28]

    C. T. Campana, F. Topputo, Clustering of earth-moon transfers in sun-perturbed environment, in: 2024 IEEE Congress on Evolutionary Computation (CEC), IEEE, 2024, pp. 1–8.doi:10.1109/CEC60901.2024.10611828

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1109/cec60901.2024.10611828 2024

  29. [29]

    Pinelli, Neural network applications to bi-impulsive earth-moon transfers in a four-body model, Master’s thesis, Politecnico di Milano (2023)

    A. Pinelli, Neural network applications to bi-impulsive earth-moon transfers in a four-body model, Master’s thesis, Politecnico di Milano (2023)

    2023

  30. [30]

    Litteri, A

    W. Litteri, A. Francisco Gil, M. Vasile, V. Rodriguez-Fernandez, D. Camacho, Generation of periodic orbits in the restricted three-body problem with a variational autoencoder, Celest. Mech. Dyn. Astron. 138 (2026) 25.doi:10.1007/s10569-026-10299-x

    work page doi:10.1007/s10569-026-10299-x 2026

  31. [31]

    L. Dinh, J. Sohl-Dickstein, S. Bengio, Density estimation using real nvp, arXiv preprint arXiv:1605.08803 (2016).doi:10.48550/arXiv. 1605.08803

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

  32. [32]

    J.Graebner,R.Beeson,Globalsearchforoptimallowthrustspacecrafttrajectoriesusingdiffusionmodelsandtheindirectmethod,J.Astronaut. Sci. 72 (6) (2025) 62.doi:10.1007/s40295-025-00535-1

    work page doi:10.1007/s40295-025-00535-1 2025

  33. [33]

    T.Presser,A.Dasgupta,D.Erwin,A.Oberai,Diffusionmodelsforgeneratingballisticspacecrafttrajectories,arXivpreprintarXiv:2405.11738 (2024).doi:10.48550/arXiv.2405.11738

    work page doi:10.48550/arxiv.2405.11738 2024

  34. [34]

    S. Fu, D. Wu, S. Gong, Discontinuous behavior of time-of-flight distribution for bi-impulsive earth-moon transfers in the three-body model, arXiv preprint arXiv:2510.26567 (2025)

    arXiv 2025

  35. [35]

    J. Ho, A. Jain, P. Abbeel, Denoising diffusion probabilistic models, in: H. Larochelle, M. Ranzato, R. Hadsell, M. Balcan, H. Lin (Eds.), Advances in Neural Information Processing Systems, Vol. 33, Curran Associates, Inc., 2020, pp. 6840–6851

    2020

  36. [36]

    Vaswani, N

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. u. Kaiser, I. Polosukhin, Attention is all you need, in: I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, R. Garnett (Eds.), Advances in Neural Information Processing Systems, Vol. 30, Curran Associates, Inc., 2017. CRediT authorship contribution statemen...

    2017