Pith. sign in

REVIEW 5 major objections 5 minor 23 references

Drift-Aware Multi-Target Space Tug Logistics Using Natural Orbital Precession

T0 review · 5 major / 5 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read By shaping an intermediate orbit's inclination, eccentricity, and altitude, a spacecraft can turn Earth's J2-induced nodal precession into a cheap passive method for closing orbital-plane gaps, allowing an analytic multi-rendezvous planner

desk verdict A genuinely useful drift-orbit method with a headline benchmark claim that is honest but conditional; treat the +1.55% as a permissive-cap, best-of-two-variants number, not a strict GTOC9 replication. read the letter →

arxiv 2607.22974 v1 pith:GDSCVAZG submitted 2026-07-25 eess.SY astro-ph.EPastro-ph.IMcs.SYmath-phmath.MP

Omer Burak Iskender , Leonard Felicetti , Kaan Gokturk , David Tellett , Adam Baker This is my paper · ORCID
classification eess.SYastro-ph.EPastro-ph.IMcs.SYmath-phmath.MP MSC 70M2090C59
keywords J2precessionRAANdriftorbitmulti-rendezvousactivedebrisremovalspacetugtrajectoryoptimizationGTOC9
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

The paper tries to establish that the nodal precession caused by Earth's oblateness (the J2 perturbation) is not just something to cancel but a resource that can be tuned to save propellant in multi-target debris-removal missions. Its central claim is that by optimizing an intermediate 'drift orbit' not only in altitude but also in eccentricity and particularly inclination, a spacecraft can amplify the differential precession rate several-fold and passively close the orbital-plane (RAAN) gap between debris targets. On top of that, it claims that a global per-leg time-budget optimizer resolves the coupling between drift time and downstream geometry, and that the full 123-debris campaign can be planned analytically to within 1.55% of the published reference cost in about 27 minutes on a single CPU core. A sympathetic reader would care because this suggests that the expensive, massive precomputed transfer databases used by prior winning solutions may be unnecessary for many mission-design purposes, making rapid replanning and interactive trade studies practical.

What carries the argument

The key machinery is the three-parameter drift orbit (C+): an intermediate orbit defined by simultaneously optimized semi-major axis, eccentricity, and inclination, with the inclination as the dominant lever in the Sun-synchronous regime because |cos i| is small there, so a 2-degree inclination change alters the nodal precession rate by about 25%. A second central element is the global per-leg time-budget optimizer, a two-phase metaheuristic (differential evolution followed by coordinate descent) that allocates waiting/drift time across all legs of a mission together, accounting for how each leg's elapsed time shifts the RAAN geometry of every later leg; this resolves the coupling that defea

What would settle it

Re-run all ten missions of the GTOC9 campaign with the per-leg transfer segment capped at 25 days (the strict arrival-to-arrival reading of the 30-day budget after subtracting the 5-day dwell) and compare the resulting total cost to the published reference of 731.28 MEUR. The paper reports only a single-mission result under that cap (Mission 10 rises to +7.2% above the reference); if the full-campaign total under the strict cap rises by more than a few percentage points above 742.65 MEUR, the 'closely matches the benchmark' claim would not survive the strict competition rules.

Watch

Extended reading notes

Core claim

The paper's central claim is that J2-induced nodal precession, normally treated as a perturbation to be cancelled, can be engineered as a cheap propulsive resource: by choosing an intermediate drift orbit's semi-major axis, eccentricity, and especially inclination, a spacecraft amplifies the differential RAAN drift rate several-fold and closes the orbital-plane gap between debris targets passively during coasting. On the paper's own terms, an analytic pipeline using two-impulse Hohmann transfers plus a global per-leg time-budget optimizer reproduces the published GTOC9 reference campaign cost (731.28 MEUR) to within 1.55% (742.65 MEUR) across all 123 debris targets, using minutes of single-c

Load-bearing premise

The headline full-campaign cost match depends on reading the competition's per-leg cap as 30 days for the transfer segment with the mandatory 5-day dwell added separately, which is up to five days more permissive per leg than the strict arrival-to-arrival competition reading; the paper does not report a full-campaign number under the stricter 25-day transfer cap.

Editorial extensions

If this is right

  • If the central claim is correct, multi-rendezvous debris removal and in-orbit servicing campaigns can be designed and re-planned in minutes on a laptop, without the multi-hundred-million-row transfer databases used by prior reference solutions.
  • The inclination lever implies that even modest plane changes during drift-orbit insertion can create large nodal-rate amplification, potentially reducing propellant cost on moderate-RAAN-gap legs by up to about 40% compared with altitude-only shaping.
  • The global time-budget optimizer makes explicit that per-leg drift times are not independent: a greedy allocation can be catastrophically suboptimal (the paper quotes a case where it yields 10,512 m/s versus 1,434 m/s after global optimization), so any future mission planner should treat the per-leg time vector as a coupled global decision.
  • The sequence-recovery result suggests that the optimal removal order is largely dictated by J2 precession geometry, so the combinatorial sequencing subproblem can be guided by simple geometric heuristics that use accumulated elapsed time to propagate RAANs.
  • The measured +7.7% executable overhead and the 17% tail of wide-gap legs indicate that the analytic ledger is a planning approximation, not a literal execution cost; the paper argues this gap is where numerical refinement retains an advantage.

Reading between the lines

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

  • Editorial inference: the same inclination-shaping idea likely transfers beyond Sun-synchronous orbits to any regime where |cos i| is small or where small inclination changes can produce disproportionately large nodal-rate changes, potentially including critically inclined or near-polar orbits.
  • Editorial inference: the paper's own disclosure of the +3.8% nodal-rate offset as a reference-frame effect suggests that a simple epoch-dependent correction could be folded into the analytic rate model, removing that small systematic bias at negligible cost.
  • Editorial inference: because the paper reports only a Mission 10 result under the strict 25-day arrival-to-arrival cap, a natural testable extension is to re-run all ten missions under that strict cap; the outcome would determine whether the 'closely matches' claim survives the strict competition reading.
  • Editorial inference: the paper's analytic pipeline could be used as a warm-start or surrogate model inside a higher-fidelity numerical optimizer, since it identifies promising drift-orbit and time-budget regions in milliseconds, and the wide-gap tail could then be refined with multi-impulse targeting.
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

5 major / 5 minor

Summary. The paper proposes an analytical, J2-drift-aware trajectory optimization framework for multi-rendezvous debris-removal missions, benchmarked against the GTOC9 Kessler Run problem and the published JPL solution. Three contributions are claimed: (1) a three-parameter drift orbit (C+) that co-optimizes semimajor axis, eccentricity, and inclination; (2) a two-phase per-mission global time-budget optimizer that resolves the coupling between per-leg drift duration and downstream RAAN geometry; (3) an independent sequence-recovery test on JPL Mission 10. The headline result is a full-campaign cost of 742.65 MEUR on the published JPL 10-mission partition, against 731.28 MEUR for the JPL reference (+1.55%), obtained with analytic Hohmann-type transfers and about 27 minutes of single-core computation. The paper also reports a GMAT-anchored numerical validation of the analytic transfer model and a measured +7.7% executable-decomposition overhead.

Significance. If the headline comparison were fully supported, the paper would be a significant methodological result: a lightweight analytic pipeline reproducing the cost level of a solution that used a ~290-million-row semi-analytic transfer database and SNOPT refinement. The paper is unusually transparent about several limitations: Sec. 2.3 discloses that the runs labelled 'GTOC9-compliant' use a more permissive reading of the per-leg time cap; Sec. 6.4 reports the executable overhead, the wide-gap tail, and the residual against JPL on Mission 10; and the Fig. 1/Table 3 cost curves are explicitly identified as a calibrated illustrative model rather than optimiser outputs. These disclosures are a strength. However, the central claim as stated in the abstract is not yet established: the headline uses a per-mission best-of-greedy/global selection and a 30-day transfer-segment cap, while no full-campaign result is reported under the strict competition cap. The significance of the paper would be materially improved by reporting a single-algorithm result under the strict cap.

major comments (5)
  1. [Abstract and Sec. 2.3, Eq. (4)] The abstract states that the full-campaign match is obtained 'under the competition's transfer-time rules.' Sec. 2.3 explicitly says the opposite: the runs labelled GTOC9-compliant apply the 30-day cap to the transfer segment and add the >=5-day dwell separately, allowing up to 35 days arrival-to-arrival, whereas the strict GTOC9 cap is 25 days of transfer inside a 30-day arrival-to-arrival window. The paper reports only one strict-cap data point (Mission 10 rising to 1525 m/s, +7.2%) and provides no full-campaign total under the strict cap. Because Missions 1-3 already sit 6.8-12.5% above the JPL reference under the permissive cap, the aggregate +1.55% cannot be assumed to survive the stricter feasible set. The abstract's 'closely matches under the competition's transfer-time rules' is therefore not supported; at minimum it must be re-quantified as conditional on the more permissive cap
  2. [Table 7 and Sec. 6.2] The 742.65 MEUR headline uses the 'Best' column, which is the per-mission minimum of the greedy and global-optimised costs (M4, M5, M7 pick the greedy result; the other seven pick the global result). This is a post-hoc oracle envelope, not the output of a single algorithm. The global-only total is 748.7 MEUR, which is still close (+2.4% vs 731.28) but different from the claimed +1.55%. The paper should either report the global-only total as the framework's result or explicitly state that the 'Best' column is a selection envelope used only to bound the achievable cost, not a claim for a single pipeline.
  3. [Fig. 1 and Table 3] The large savings attributed to C+ in Fig. 1 and Table 3 (e.g. 'C+ only ~136 m/s, a 90% saving over A'; 'C+ over B: -88%') are computed from an assumed effective-rate model with 2x/5x amplification and fixed +30/+40 m/s insertion overhead, calibrated against the optimiser rather than produced by it. The caption correctly discloses this, but the surrounding text and the 'Saving of analytical C+' line in Table 3 can easily be read as optimiser-derived results. Since Sec. 6.1 reports the actually-optimised C+ benefit as ~35-53% on the most favourable Mission-10 legs, the Fig. 1/Table 3 percentages should be presented strictly as an illustrative model or removed from the results narrative.
  4. [Sec. 2.1 / Sec. 6.2] The full-campaign evaluation adopts JPL's published mission partition and JPL's mission start epochs verbatim. This is a reasonable methodological choice for isolating the trajectory layer, and the paper says so. But it means the 'framework' claim is not an end-to-end solution: the combinatorial set-cover decision, which the introduction identifies as a central part of the GTOC9 problem, is imported from the reference. The paper should explicitly state in the conclusion and abstract that the +1.55% figure is for a fixed partition, not for autonomous campaign discovery. This is not a fatal flaw, but it is essential to prevent the result from being over-interpreted.
  5. [Sec. 6.4, 'Wide-gap tail' and 'Implications'] The numerical validation shows that 15 of 89 verified C+ legs miss the target plane by a median of 5.7 degrees in RAAN, and the paper states that closing these legs requires a dedicated plane-targeting burn whose cost is not quantified and could be of order 1 km/s on the worst leg, potentially pushing literal execution against the 5000-kg tank capacity on Missions 1-3. The paper is honest that these are extrapolations. However, the conclusion retains the '+1.55%' as the principal result without incorporating these caveats. The conclusion should state that the analytic-ledger cost is a planning-level estimate and that the unquantified wide-gap tail is a material uncertainty for the campaign-level match.
minor comments (5)
  1. [Fig. 5 caption] The caption reports 'greedy 869, global 745, best-per-mission 739, JPL 732 MEUR', but Table 7 reports greedy 872.7, global 748.7, best 742.65, JPL 731.5 MEUR. These numbers should be reconciled.
  2. [Abstract / Sec. 2.3] The phrase 'GTOC9-compliant' is used throughout for runs that the paper itself defines as more permissive than the competition rule. Consider using 'GTOC9-referenced' or 'GTOC9-inspired' to avoid mislabelling.
  3. [Sec. 6.1, sequence recovery] The independent sequence-recovery validation uses only Method B (wait-for-drift) under a 16-day cap. It would be useful to state that C+ and the global time-budget optimizer were not exercised in this recovery test.
  4. [Sec. 4.2 / Sec. 6.1] The introduction claims the global time-budget optimizer 'closes [the sequence-timeline coupling] by up to an order of magnitude.' The reported numbers show a ~2.6x improvement under the 30-day cap (3906 to 1484 m/s) and ~7.3x under unconstrained drift (10512 to 1434 m/s), the latter being a sensitivity case that violates the cap. The 'order of magnitude' claim should be qualified or sourced to the companion paper.
  5. [Table 1] Table 1 lists T_max = 30 d as the per-leg transfer-time cap, while Sec. 2.3 explains that the strict competition reading permits only 25 d of transfer within a 30-day arrival-to-arrival window. The table should distinguish these two quantities.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity; the central analytic pipeline is derived from J2 equations and benchmarked against an external reference. The disclosed 30-day-cap reading and the explicitly calibrated illustrative model affect interpretation, but are not hidden reductions.

full rationale

The core transfer-cost ledger derives from the standard J2 secular-rate equation (12) and the GTOC9 cost function (1)-(3); the per-mission optimizer (Sec. 4) operates on those equations rather than fitting to JPL's 731.28 MEUR. The main validation (Sec. 6.2, Table 7) compares an analytic pipeline to an external reference, so the +1.55% figure is not constructed by definition. The manuscript itself flags the only apparent reduction: Fig. 1/Table 3's effective-rate model is 'calibrated against the optimiser rather than produced by it' and its '~5x amplification' and '~40 m/s insertion' are 'model inputs, not findings'; the magenta-star optimizer output is the only derived point. This is an explicitly disclosed illustrative device, not a disguised prediction. Self-citations [14],[15] delegate supplementary verification and formal problem statements to same-author companions, but the paper's numerical results are generated in-house and benchmarked to the independent JPL/GTOC9 data, so the self-citation is not load-bearing. The remaining caveat—the 'GTOC9-compliant' runs use a 30-day transfer-segment cap with the dwell added separately, up to five days more permissive than the strict arrival-to-arrival reading, and no strict-cap full-campaign total is reported—is a rule-interpretation/validity concern, not circularity: the per-leg costs still follow from the J2 drift equations rather than from the benchmark total. Similarly, Table 7's 'Best' column chooses the lower of greedy/global per mission, so 742.65 MEUR is an envelope rather than a single fixed-algorithm output; this weakens the strength of the match claim but is not a reduction to inputs. Score 2 reflects the minor self-citation and the disclosed calibration, not a circular derivation.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

No new physical entities are invented; the drift orbit is a trajectory design, not a new object. The free parameters are optimisation hyperparameters and the explicitly calibrated illustrative model; none enter the central J2 derivation as fitted physical constants.

free parameters (4)
  • Effective-rate amplification factors and insertion overhead = 2×/5× rate amplification; +30/+40 m/s insertion overhead
    Used to draw the A/B/C/C+ cost curves in Fig. 1 and Table 3. The paper states these are 'calibrated against (not produced by) the optimiser' and are 'model inputs, not findings.'
  • Drift-orbit grid bounds = e_d ∈ [0,0.06]; i_d ∈ [i_src−3°, i_src+3°]
    Hand-chosen search domain for the C+ two-dimensional grid search; affects achievable solutions.
  • Coordinate-descent time grid = {0.5, 1, 2, 5, 10, 16, 20, 30, 45, 60} days
    Discrete T_k grid for the refinement phase; chosen by hand.
  • Differential evolution hyperparameters = population 20/15; F=0.5; CR=0.7; maxiter 100/60; seed 42
    Metaheuristic settings chosen by the authors. Sec. 6.1 notes different configurations give different results, so these choices affect the reported numbers.
assumptions (5)
  • standard math J2 secular precession rates (Eqs. 12–13) with linear propagation of Ω and ω.
    Standard first-order J2 secular theory, taken from the GTOC9 specification [5]; the transfer-cost model consumes only Ωdot.
  • domain assumption Two-impulse Hohmann transfers with combined plane change are the appropriate cost model for the considered transfer methods.
    Used throughout to build the A/B/C/C+ cost ledger (Sec. 4.1); explicitly set aside multi-impulse refinement benefits.
  • domain assumption Argument of perigee and mean anomaly need not be propagated in the cost model.
    Sec. 2.6 and Remark 1 rely on the GTOC9 waiver; the paper acknowledges the consequences are only characterised empirically.
  • domain assumption Circular-orbit assumption for transfer-cost evaluation despite catalogue eccentricities up to e≈0.02.
    Sec. 7 Limitations; the paper lists this as a source of residual vs. the JPL reference.
  • domain assumption The published JPL mission partition is a valid controlled benchmark for isolating the trajectory layer.
    Adopted in Sec. 2.4 and 6.2; if the partition embeds JPL's transfer-model assumptions, the comparison is not an independent end-to-end validation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Drift-Aware Multi-Target Space Tug Logistics Using Natural Orbital Precession." pith.science (2026). https://pith.science/paper/GDSCVAZG

@misc{pith2026260722974,
  author       = {Pith},
  title        = {Pith review of: Drift-Aware Multi-Target Space Tug Logistics Using Natural Orbital Precession},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GDSCVAZG}},
  note         = {Machine review of arXiv:2607.22974}
}
read the original abstract

Multi-rendezvous missions for active debris removal and in-orbit servicing incur prohibitive propellant costs when their targets differ in the orientation of their orbital planes, and classical mission design treats the nodal precession induced by Earth's oblateness as a perturbation to be cancelled. This paper presents a drift-aware trajectory optimisation framework that instead exploits that precession as a mission-design resource, benchmarked against the published winning solution to the European Space Agency's Kessler Run trajectory-optimisation competition as a trusted reference. Three methodological contributions are made. First, an enhanced drift orbit shapes the size, shape, and inclination of an intermediate orbit to tune the differential precession rate, substantially cheaper than the altitude-only designs of prior work in the Sun-synchronous regime, where inclination is the dominant lever. Second, a global per-leg time-budget optimiser makes explicit and resolves the coupling between drift-time allocation and downstream nodal geometry that defeats greedy allocation. Third, a sequence-recovery test independently re-derives a removal order consistent with the published reference from the debris catalogue alone. Applied to the full debris campaign on the published mission partition under the competition's transfer-time rules, the framework closely matches the published benchmark cost using only analytical transfer models and minutes of single-core computation. A cross-validation anchored to a widely used flight-dynamics tool quantifies the fidelity limits of the analytic model and bounds the cost of literal execution, and the constructive nature of the algorithm supports rapid in-orbit replanning after a missed manoeuvre.

Figures

Figures reproduced from arXiv: 2607.22974 by the authors.

Figure 1
Figure 1. Three-parameter drift-orbit concept on JPL Mission 7 Leg 6 (GTOC9 D46 [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Drift-orbit element comparison for M7 L6 (D46 [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Per-leg Δ𝑉 on the published JPL Mission 10 reference sequence, with labels above each hybrid bar marking the selected sub-method (A, B, or C+ ). Bars are shown alongside the published JPL reference values for direct visual comparison. 10 4 Mission ¢V (m/s, log scale) Unconstrained (sensitivity only) ¢tR 30 d/leg (GTOC9-compliant) 16 d/leg 10,512 1,434 3,906 1,484 1,669 1,682 JPL 1,423 m/s +4.3% vs JPL Global vs gree… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Reduction ladder for the global time-budget op [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]
Figure 5
Figure 5. Figure 5: Full 123-debris campaign benchmark context: per-mission cost [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 6
Figure 6. Figure 6: Cumulative correlation deviation 1−𝑟(𝑡) vs. elapsed time for the seven state channels (3 position, 3 velocity, longitude), FPROX vs. GMAT. Vertical dotted lines mark the Day 1, Day 30, and Day 90 checkpoints. All channels stay below 10−4 throughout the 90-day propagati…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

23 extracted references

  1. [1]

    DISCOSweb – Space debris statistics.https://sdup.esoc.esa.int/ discosweb/statistics/, 2026

    European Space Agency. DISCOSweb – Space debris statistics.https://sdup.esoc.esa.int/ discosweb/statistics/, 2026. accessed 2026

    2026

  2. [2]

    D. J. Kessler and B. G. Cour-Palais. Collision frequency of artificial satellites: The creation of a debris belt.Journal of Geophysical Research, 83(A6):2637–2646, 1978

    1978

  3. [3]

    Dario Izzo, Ingmar Getzner, Daniel Hennes, and Luís F. Simões. Evolving solutions to TSP variants foractivespacedebrisremoval. InProceedingsofthe Genetic and Evolutionary Computation Conference, pages 1207–1214. ACM, 2015

    2015

  4. [4]

    GTOC9: The Kessler Run.https://kelvins.esa.int/ gtoc9-kessler-run/, 2017

    European Space Agency. GTOC9: The Kessler Run.https://kelvins.esa.int/ gtoc9-kessler-run/, 2017

    2017

  5. [5]

    Izzo and M

    D. Izzo and M. Märtens. The Kessler Run: On the designoftheGTOC9challenge.ActaFutura,11:11– 24, 2018

    2018

  6. [6]

    Petropoulos, D

    A. Petropoulos, D. Grebow, D. Jones, G. Lantoine, A. Nicholas, J. Roa, J. Senent, J. Stuart, N. Arora, T.Pavlak,T.Lam,T.McElrath,R.Roncoli,D.Garza, N. Bradley, D. Landau, Z. Tarzi, F. Laipert, E. Bon- figlio, M. Wallace, and J. Sims. GTOC9: Methods and results from the Jet Propulsion Laboratory team. Acta Futura, 11, 2018. 15

    2018

  7. [7]

    Dorigo and L

    M. Dorigo and L. M. Gambardella. Ant colony sys- tem: Acooperativelearningapproachtothetraveling salesmanproblem.IEEETransactionsonEvolution- ary Computation, 1(1):53–66, 1997

    1997

  8. [8]

    P. E. Gill, W. Murray, and M. A. Saunders. SNOPT: An SQP algorithm for large-scale constrained opti- mization.SIAM Journal on Optimization, 12:979– 1006, 2002

    2002

Show all 23 references
  1. [9]

    Y. Luo, Y. Zhu, H. Zhu, Z. Yang, S. Mou, J. Zhang, Z. Sun, and J. Liang. GTOC9: Results from the National University of Defense Technology.Acta Futura, 11, 2018

    2018

  2. [10]

    H. Li, S. Chen, H. Yang, Z. Chi, G. Tang, F. Jiang, and H. Baoyin. GTOC9: Results and methods of team2–TsinghuaUniversity.ActaFutura,11,2018

    2018

  3. [11]

    Hallmann, M

    M. Hallmann, M. Schlotterer, A. Heidecker, M. Sagliano, F. Fumenti, V. Maiwald, and R. Schwarz. GTOC 9, multiple space debris ren- dezvous trajectory design in the J2 environment. Technical report, Institute of Space Systems, DLR, Bremen, Germany, 2017

    2017

  4. [12]

    C. O. Absil, L. A. Ricciardi, M. Di Carlo, C. Greco, R. Serra, M. Polnik, A. Vroom, A. Riccardi, E. Minisci, and M. Vasile. GTOC 9: Results from the University of Strathclyde (team Strath++).Acta Futura, 11:57–70, 2018

    2018

  5. [13]

    L. F. Simões, D. Izzo, E. Haasdijk, and A. E. Eiben. Multi-rendezvous spacecraft trajectory op- timization with beam P-ACO. InEvolutionary Com- putation in Combinatorial Optimization, pages 141–

  6. [14]

    O. B. Iskender, K. Gokturk, D. Tellett, and A. Baker. Turning perturbations into propellant: Drift-aware multi-rendezvous trajectory optimization for in-orbit servicing.Acta Astronautica, 2026

    2026

  7. [15]

    Space postman and ATLAS: Enabling startup participation in the orbital logistics economy

    Iskender, Omer Burak. Space postman and ATLAS: Enabling startup participation in the orbital logistics economy. InProc.77thInt.AstronauticalCongr.Int. Astronautical Federation, Antalya, Türkiye, 2026. PaperIAC-26,E6,1,10,x115070, IAFBusinessesand Innovation Symposium

    2026

  8. [16]

    Dual quaternion based autonomous rendezvous and dock- ing via model predictive control

    Omer Burak Iskender, Keck-Voon Ling, Luca Si- monini, Markus Schlotterer, David Seelbinder, Stephan Theil, and Jan Marian Maciejowski. Dual quaternion based autonomous rendezvous and dock- ing via model predictive control. InProc. 70th Int. Astronautical Congr. Int. Astronauti...

    2019

  9. [17]

    PhD thesis, Nanyang Technological University, Singapore, 2020

    Omer Burak Iskender.Model predictive control for spacecraft rendezvous and docking with uncooper- ative targets. PhD thesis, Nanyang Technological University, Singapore, 2020

    2020

  10. [18]

    Reducibilityamongcombinatorialprob- lems

    R.M.Karp. Reducibilityamongcombinatorialprob- lems. InComplexity of Computer Computations, pages 85–103. Springer, 1972

    1972

  11. [19]

    Differentialevolution—asim- pleandefficientheuristicforglobaloptimizationover continuous spaces.Journal of Global Optimization, 11(4):341–359, 1997

    R.StornandK.Price. Differentialevolution—asim- pleandefficientheuristicforglobaloptimizationover continuous spaces.Journal of Global Optimization, 11(4):341–359, 1997

    1997

  12. [20]

    Gamrath et al

    G. Gamrath et al. The SCIP optimization suite 3.2. TechnicalReportTechnicalReport15-60,ZIB,2016

    2016

  13. [21]

    InProceedingsof theTwenty-FourthInternationalJointConferenceon Artificial Intelligence (IJCAI), pages 769–775, 2015

    D.HennesandD.Izzo.Interplanetarytrajectoryplan- ningwithMonteCarlotreesearch. InProceedingsof theTwenty-FourthInternationalJointConferenceon Artificial Intelligence (IJCAI), pages 769–775, 2015

    2015

  14. [22]

    Biscani and D

    F. Biscani and D. Izzo. A parallel global multiobjec- tive framework for optimization: pagmo.Journal of Open Source Software, 5(53):2338, 2020

    2020

  15. [23]

    Soft- ware, available athttps://software.nasa.gov/ software/GSC-17177-1

    NASA Goddard Space Flight Center.General Mis- sion Analysis Tool (GMAT) R2025a, 2025. Soft- ware, available athttps://software.nasa.gov/ software/GSC-17177-1. 16

    2025

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.