arXiv:2605.26790v1 Announce Type: new Abstract: Low-thrust trajectory design relies heavily on repeated evaluations of fuel consumption and transfer feasibility, which require expensive optimal control solutions. In this work, we show these quantities can be accurately approximated by machine learning surrogates, enabling fast and scalable evaluation across a wide range of scenarios. By increasing both dataset size and model capacity, we observe that low-thrust trajectory optimization follows a scaling law, with performance improving linearly with the logarithm of training data and network par
The increasing complexity and computational demands of low-thrust trajectory optimization, coupled with advancements in machine learning, are creating a ripe environment for AI-driven solutions.
This development could dramatically reduce the cost and time required for designing and evaluating space missions, enabling more ambitious and frequent exploration and commercial activities.
The expensive, iterative process of trajectory design can now be significantly accelerated and scaled, using AI to rapidly approximate optimal solutions. This could democratise space access beyond the major space powers.
- · Space agencies
- · Private space companies
- · AI/ML model developers
- · Satellite operators
- · Traditional aerospace trajectory optimization specialists (without ML skills)
- · Companies reliant on bespoke, long-lead trajectory design
Faster and cheaper low-thrust trajectory design for various space missions.
Increased accessibility and efficiency in deploying constellations, long-duration missions, and complex orbital maneuvers.
Potential for new space-based services and capabilities that were previously economically or computationally unfeasible, accelerating the space economy and defence capabilities to a greater extent, as space based defence assets could be moved more quickly and at greater scale/reduced cost. This also suggests more satellites and space debris generally as mass is moved around the solar system.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG