Project CERES: Nuclear Thermal Transportation and In-Situ Propellant Production A Foundational Architecture for Solar System Logistics

arXiv:2609.40011 · astro-ph.EP, astro-ph.IM · Submitted 2026-09-30 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.

Vera: I'm Vera, and with me are Jocelyn and Subrahmanyan, guest researcher.

Jocelyn: Today's paper: "Project CERES: Nuclear Thermal Transportation and In-Situ Propellant Production A Foundational Architecture for Solar System Logistics".

Vera: Project CERES proposes a reusable deep-space transportation architecture centered on Ceres, combining nuclear thermal propulsion (NTP) with in-situ production of water, hydrogen, oxygen and ammonia.

Jocelyn: First, who's behind it and why it matters.

Title and authors: Vera: So, to get us started, the paper outlines several key areas where advanced AI could be used to manage the complexity of running this entire system efficiently over time.

Jocelyn: It seems they are really zeroing in on optimizing how resources are extracted at Ceres, specifically trying to balance the rate of water and other chemical production against exactly what is needed for scheduled return trips.

Subrahmanyan: That scheduling optimization is critical because it moves us from a static plan to a dynamic model that can react to real-time fluctuations in supply on the ground, which is a big step forward.

Vera: They also suggest an AI system for autonomous node operations, which would handle coordinating all those ferry turnaround times and managing the inventory of various propellant storage systems efficiently.

Jocelyn: I think that level of autonomy is substantial; having an AI manage those schedules ensures that the node adheres strictly to its timetable, no matter what external factors affect mission arrival or resource availability.

Subrahmanyan: That system bridges the gap between theoretical capability and practical execution by modeling how different operational limits, like storage capacity for ammonia versus hydrogen, interact in a dynamic way.

Vera: Another suggestion is an AI designed for autonomous vehicle trajectory planning, which would specifically focus on optimizing those departure vectors from Ceres to maximize efficiency during the transfer legs.

Jocelyn: That ties back directly to the transfer problem we talked about earlier; this AI would be designed to find those most repeatable windows and make sure every mission takes advantage of them optimally.

Subrahmanyan: By focusing trajectory planning around that "node effect," they are essentially using Ceres’ position and gravity as a strategic asset instead of just treating it as some random waypoint for a spacecraft.

Vera: These improvements really focus on making the entire hub-and-spoke network truly self-regulating, where the scheduling and resource allocation happen automatically based on what's happening right now.

Jocelyn: It seems like they are moving beyond just proving the concept to designing an actual operational framework that could be run by sophisticated systems before we even start building the physical hardware itself.

Subrahmanyan: The economic modeling aspect is also important, suggesting how these AI tools can help project whether establishing a permanent node is economically sensible based on how variable the resource grades at Ceres might actually be.

Vera: So, they are not just showing us what’s possible with current technology, but they are mapping out exactly what the next generation of systems needs to do to make this concept a working reality.

Jocelyn: It’s exciting because it shows that we can tackle problems involving complex resource management and dynamic scheduling using advanced computational methods before we even commit to designing the physical hardware itself.

Subrahmanyan: The future work they propose points toward integrating these AI tools directly into the mission planning pipeline, which could potentially alter how we approach long-duration space infrastructure projects entirely.

The paper's summary: Vera: So, to wrap up this discussion on "Project CERES: Nuclear Thermal Transportation and In-Situ Propellant Production A Foundational Architecture for Solar System Logistics," we’ve seen how the authors suggest specific AI enhancements to manage the complexity of operating this whole system efficiently.

Jocelyn: It really shows that combining nuclear thermal propulsion with local resource production isn't just an interesting theoretical exercise; it’s a concrete way to approach building infrastructure in the solar system systematically, and these AI suggestions help make that systematic approach work.

Subrahmanyan: The long-term implication is significant for our understanding of solar system logistics, suggesting a pathway toward continuous operations rather than those sporadic, high-energy expeditions we usually see.

Vera: Exactly; we're looking at a future where we can establish more permanent outposts by utilizing what’s available in the asteroid belt itself through these kinds of intelligent operational layers.

Jocelyn: And with the detailed analysis they present on transfer windows and propulsion trade-offs, it gives us a much clearer picture of what is actually achievable in the near term for these kinds of missions.

Subrahmanyan: That work connects fundamental physics—like specific impulse ranges and C3 requirements—directly to practical engineering constraints, which helps us define the boundaries of what is physically feasible for these architectures.

Vera: It feels like we’ve seen a real roadmap for how we might transition deep space exploration from just sending things out and back to running a kind of interplanetary supply chain powered by intelligent management.

Jocelyn: The idea that Ceres can support three different propellant chains simultaneously is a really smart feature for flexibility in future mission planning, and the AI tools help us utilize that flexibility fully.

Subrahmanyan: Ultimately, Project CERES demonstrates how integrating resource utilization and propulsion design can manage mass and time constraints together across multiple mission phases efficiently through these computational aids.

The paper's improvements: Vera: So we've just gone through the Project CERES paper, which lays out this complete blueprint for using Ceres as a central logistics hub with nuclear thermal propulsion and in-situ propellant production.

Jocelyn: It really shows us how combining those two technologies can create a way to approach building infrastructure in the solar system systematically, moving beyond just sending things out and back.

Subrahmanyan: The long-term implication is significant for our understanding of solar system logistics, suggesting a pathway toward continuous operations instead of those sporadic, high-energy expeditions we usually see.

Vera: Exactly; we're looking at a future where we can establish more permanent outposts by utilizing what’s available in the asteroid belt itself.

Jocelyn: And with the detailed analysis they present on transfer windows and propulsion trade-offs, it gives us a much clearer picture of what’s actually achievable in the near term for these kinds of missions.

Subrahmanyan: That work connects fundamental physics—like specific impulse ranges and C3 requirements directly to practical engineering constraints, which helps us define the boundaries of what is feasible.

Vera: It feels like we’ve seen a real roadmap for how we might transition deep space exploration from just sending things out and back to running a kind of interplanetary supply chain.

Jocelyn: The idea that Ceres can support three different propellant chains simultaneously is a really smart move for flexibility in the future mission planning.

Subrahmanyan: Ultimately, Project CERES demonstrates how integrating resource utilization and propulsion design can manage mass and time constraints together across multiple mission phases efficiently through these computational aids.

Vera: We’ve seen how this paper lays out a comprehensive vision for a sustainable deep space network centered around Ceres, which is really significant for the future of human activity in the solar system.

Jocelyn: It really shows us that combining nuclear thermal propulsion with local resource production isn't just an interesting theoretical exercise; it’s a concrete way to approach building infrastructure in the solar system.

Subrahmanyan: The long-term implication is significant for our understanding of solar system logistics, suggesting a pathway toward continuous operations rather than those sporadic, high-energy expeditions.

Vera: Exactly; we're looking at a future where we can establish more permanent outposts by utilizing what’s available in the asteroid belt itself.

Jocelyn: And with the detailed analysis they present on transfer windows and propulsion trade-offs, it gives us a much clearer picture of what’s actually achievable in the near term for these kinds of missions.

Subrahmanyan: That work connects fundamental physics—like specific impulse ranges and C3 requirements directly to practical engineering constraints, which helps us define the boundaries of what is feasible.

Vera: It feels like we’ve seen a real roadmap for how we might transition deep space exploration from just sending things out and back to running a kind of interplanetary supply chain.

Jocelyn: The idea that Ceres can support three different propellant chains simultaneously is a really smart move for flexibility in the future mission planning.

Subrahmanyan: Ultimately, Project CERES demonstrates how integrating resource utilization and propulsion design can manage mass and time constraints together across multiple mission phases efficiently through these computational aids.

Conclusion: Vera: So we've just gone through the Project CERES paper, which lays out a complete blueprint for using Ceres as a central logistics hub with nuclear thermal propulsion and in-situ propellant production.

Jocelyn: It really shows us how combining those two technologies can create a way to approach building infrastructure in the solar system systematically, moving beyond just sending things out and back.

Subrahmanyan: The long-term implication is significant for our understanding of solar system logistics, suggesting a pathway toward continuous operations rather than those sporadic, high-energy expeditions.

Vera: Exactly; we're looking at a future where we can establish more permanent outposts by utilizing what’s available in the asteroid belt itself.

Jocelyn: And with the detailed analysis they present on transfer windows and propulsion trade-offs, it gives us a much clearer picture of what’s actually achievable in the near term for these kinds of missions.

Subrahmanyan: That work connects fundamental physics—like specific impulse ranges and C3 requirements—directly to practical engineering constraints, which helps us define the boundaries of what's feasible.

Vera: It feels like we’ve seen a real roadmap for how we might transition deep space exploration from just sending things out and back to running a kind of interplanetary supply chain.

Jocelyn: The idea that Ceres can support three different propellant chains simultaneously is a really smart move for flexibility in the future mission planning.

Subrahmanyan: Ultimately, Project CERES demonstrates how integrating resource utilization and propulsion design can manage mass and time constraints together across multiple mission phases efficiently.

Vera: So, to wrap up this discussion on "Project CERES: Nuclear Thermal Transportation and In-Situ Propellant Production A Foundational Architecture for Solar System Logistics," we’ve seen how this paper lays out a comprehensive vision for a sustainable deep space network centered around Ceres.

Jocelyn: It really shows us that combining nuclear thermal propulsion with local resource production isn't just an interesting theoretical exercise; it’s a concrete way to approach building infrastructure in the solar system.

Subrahmanyan: The long-term implication is significant for our understanding of solar system logistics, suggesting a pathway toward continuous operations rather than those sporadic, high-energy expeditions.

Vera: Exactly; we're looking at a future where we can establish more permanent outposts by utilizing what’s available in the asteroid belt itself.

Jocelyn: And with the detailed analysis they present on transfer windows and propulsion trade-offs, it gives us a much clearer picture of what’s actually achievable in the near term for these kinds of missions.

Subrahmanyan: That work connects fundamental physics—like specific impulse ranges and C3 requirements—directly to practical engineering constraints, which helps us define the boundaries of what's feasible.

Vera: It feels like we’ve seen a real roadmap for how we might transition deep space exploration from just sending things out and back to running a kind of interplanetary supply chain.

Jocelyn: The idea that Ceres can support three different propellant chains simultaneously is a really smart move for flexibility in the future mission planning.

Subrahmanyan: Ultimately, Project CERES demonstrates how integrating resource utilization and propulsion design can manage mass and time constraints together across multiple mission phases efficiently.

S. P. Worden, F.G. Kennedy

astro-ph.EP, astro-ph.IM

Submitted: 2026-09-30

Updated: 2026-09-30

Comments: 14 pages, Submitted to The Journal of the British Interplanetary Society, 28 Sep 2026

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 65/100

The gist: Project CERES proposes a reusable deep-space transportation architecture centered on Ceres, combining nuclear thermal propulsion (NTP) with in-situ production of water, hydrogen, oxygen and ammonia.

Key concepts

Nuclear Thermal Propulsion (NTP)
NTP is a propulsion class that uses nuclear heat to expand a working fluid (like water) into a high-velocity exhaust. It is unique because it can efficiently close the mass, time, and return leg of deep-space travel simultaneously, unlike chemical or electric options.
Ceres as Port of Entry
Ceres is utilized as a logistics hub due to its formation from silicates and water ice. It contains water ice for direct reaction mass, ammonia precursors for industrial chemistry, and carbon sources for organics, enabling the production of multiple propellant chains.
Transportation Node
The node functions as a hub providing four services: a propellant depot, turnaround/refit facilities for vehicles not returning to Earth's gravity well, staging/assembly points, and eventual fabrication. It maintains a strict timetable for departures.
Propellant Chains
Ceres resources support three distinct propellant chains: water heated directly by NTP (specific impulse near 390 s), ammonia (a storable soft cryogen with about 500 s efficiency), and electrolytic hydrogen (requiring electrolysis and liquefaction, offering 900 s efficiency).

Terminology

Summary

Project CERES proposes a reusable deep-space transportation architecture centered on Ceres, combining nuclear thermal propulsion (NTP) with in-situ production of water, hydrogen, oxygen and ammonia. Its first flight, a Pathfinder that surveys Ceres and returns samples, also bears on the question of life: Dawn found the ingredients of prebiotic chemistry there, and returned material would show just how far that chemistry went.

The gist: Ceres uniquely combines abundant volatiles, exceptionally low gravity and a strategic Main Belt position, and that nuclear thermal propulsion is the only propulsion class that closes mass, time and the return leg together. Together they convert deep-space exploration from independent expeditions into a network operating on a timetable.

Ceres as Port of Entry

Ceres has been classified as a planet, an asteroid and a dwarf planet, motivated by its formation from silicates, water ice and volatile-rich compounds. Its subsurface ocean left hydrated silicates, carbonates, ammoniated minerals and organics before it froze into an ice-rich crust. For a transportation architecture inventory on Ceres includes water for direct reaction mass, hydrogen and hydrolox; nitrogen-bearing minerals for ammonia and industrial chemistry; and carbon in carbonates and organics. This combination of resources allows the node to support three propellant chains: water heated directly in a nuclear thermal engine (specific impulse near 390 s), ammonia (about 500 s, a storable soft cryogen), and electrolytic hydrogen (900 s, requiring electrolysis, liquefaction and zero-boiloff storage).

The Transportation Problem

Direct transits to Ceres are expensive because the required departure characteristic energy (C3) is high—43–45 km2/s—and Ceres offers neither an atmosphere for aerocapture nor a gravity well deep enough to assist propulsive capture. The oneway impulsive requirement at the most favorable windows of the 2030s is 10.2–11.2 km/s, which exceeds the total ΔV of most Earth-Mars round trips. Ceres' orbit is inclined 10.6° to the ecliptic, meaning opportunity quality depends on transfer geometry relative to the line of nodes, with per-window optima ranging from 10.2 to 14.1 km/s.

The Propulsion Trade

The paper compares chemical, electric, and nuclear thermal propulsion against a requirement to transport 35 t from low Earth orbit to low Ceres orbit. Chemical propulsion fails because it cannot hold a timetable, and a two-stage expendable stack does not close at all when charged with spacestorable propellants at 320 s. Electric propulsion closes the mass equation but loses the calendar. Nuclear thermal propulsion is unique because it is the only class that closes mass, time and the return leg together, at every synodic opportunity surveyed.

The Node and Return Leg

The architecture is a hub-and-spoke logistics network with Ceres as the hub, providing four services: a propellant depot, ferry turnaround and refit for vehicles that never return to Earth's gravity well, staging and assembly for onward legs, and eventually fabrication. The physical layout utilizes Ceres' low surface gravity of 0.028 g to make the surface-to-orbit leg nearly free; a round-trip tanker sortie costs under 1,000 m/s at a mass ratio of 1.3. The node's defining property is the timetable it keeps: departures from Earth every synodic period regardless of window quality, and departures from Ceres whenever physics permits.

Project CERES Pathfinder Mission

The Project CERES Pathfinder mission is the first operational realization of this architecture, consisting of one vehicle, launched once, that crosses the inner solar system in a year. It surveys Ceres, touches its surface to extract water from ground that has held it for four billion years, and returns samples. The mission carries a four-engine cluster and includes elements like a deployable lander for the extraction demonstration and a Stardust-class return capsule. This flight demonstrates the capability to leave Earth, replenish itself at another world, and take up station in cislunar space intact, fueled by its destination.

Economics

A transportation node justifies itself by reducing the cost of deep-space access relative to launching every mission independently from Earth. A vehicle that refuels at Ceres cuts LEO launch mass per delivery by 61% before any credit for vehicle reuse, and the launch component of delivered cost falls from a range of roughly 1,590–15,900/kg to 630–6,300/kg. The node's market is defined by propellant compatibility, as it serves nuclear-thermal vehicles on hydrogen, ammonia or steam, and chemical stages tanking locally produced hydrolox.

Improvements for AI systems

Based on the Project CERES architecture described in this paper, here are specific improvements that could be made to AI systems:

  1. Aero-Propellant Optimization AI:

  2. Aero-Propellant Optimization AI can use the detailed Lambert analysis and ephemeris data (Table 1, Figure 3) to rapidly calculate the optimal launch windows for Earth-Ceres transfers across multiple propulsion classes (Chemical, Nuclear Thermal, Nuclear Electric).

  3. It can perform real-time trade studies between propulsion types based on dynamic constraints like required Specific Impulse (Isp) versus available power/mass ratios.

  4. The improved AI system could predict the most frequent and repeatable favorable geometry windows for a given mission profile, reducing the reliance on manual ephemeris lookups and increasing mission success probability.

  5. In-Situ Resource Utilization (ISRU) Logistics AI:

  6. This AI can optimize the Phase 1 and Phase 2 buildout schedule at Ceres by balancing resource extraction rates (e.g., water vs. ammonia production) against the required propellant demand for scheduled return sorties, minimizing downtime between phases.

  7. It can model the impact of varying ore grades (as mentioned in Section 5) on the cost-benefit analysis of building a permanent node versus relying on Earth-launched supplies, providing economic breakeven projections based on launch cost regimes (Section 8).

  8. Autonomous Node Operations AI:

  9. This system can manage the hub-and-spoke logistics network by autonomously scheduling ferry turnaround times, coordinating propellant depot inventory across different propulsion systems (e.g., managing LH2 vs. ammonia storage), and optimizing the staging/assembly processes for vehicles never returning to Earth's gravity well.

  10. The improved system can dynamically adjust the operational schedule based on real-time resource availability at Ceres and scheduled vehicle arrival times, ensuring the node adheres strictly to its defined timetable rather than being subject to isolated expedition schedules.

  11. Nuclear Thermal Propulsion (NTP) System Design AI:

  12. This AI can simulate the performance envelope of different solid-core NTP engine designs (based on NERVA heritage and projected performance targets like 900 s Isp) against various temperature/fuel constraints, identifying the optimal engine configuration for a specific propellant (e.g., hydrogen vs. ammonia) and payload mass before physical hardware development begins.

  13. Autonomous Vehicle Trajectory Planning AI:

  14. This system can generate flight trajectories that explicitly incorporate the node effect—optimizing departure vectors from Ceres to maximize the Oberth advantage or minimize plane-change penalties, specifically targeting the recurring high-efficiency windows identified in Figure 3.

Abstract

Project CERES proposes a reusable deep-space transportation architecture centered on the dwarf planet Ceres, combining nuclear thermal propulsion (NTP) with in-situ production of water, hydrogen, oxygen and ammonia. Its first flight, a Pathfinder that surveys Ceres and returns samples, also bears on the question of life: Dawn found the ingredients of prebiotic chemistry there, and returned material would show just how far that chemistry went. We survey every direct Earth-Ceres opportunity from 2032 through 2042 with ephemeris-based Lambert analysis and compare chemical, gravity-assisted, nuclear-electric and nuclear thermal vehicles under one parametric mass model. Impulsive requirements vary from 10.2 to 14.1 km/s across the decade. A 900 s NTP stage delivering 35 t closes near 220 t in low Earth orbit and does so at every surveyed window; chemical delivery is marginal and sensitive to opportunity and staging, and nuclear-electric propulsion requires multi-megawatt power to match one-to-two-year trip times. Refueling at Ceres reduces the launch mass of an Earth-Ceres-Earth sortie by 61%, and the specific impulse of the propellant the node produces decides whether Ceres water becomes cargo or tanker mass. We conclude that Ceres uniquely combines abundant volatiles, exceptionally low gravity and a strategic Main Belt position, and that nuclear thermal propulsion is the only propulsion class that closes mass, time and the return leg together. Together they convert deep-space exploration from independent expeditions into a network operating on a timetable.

Related papers