The cheapest way to the Moon was apparently not the cheapest way at all.
Half a century after humanity first reached the Moon, a team of researchers in Brazil and Portugal has discovered that the cheapest known path there was never the cheapest path at all. By searching through 30 million possible trajectories — a scale previous studies never attempted — they found a route through the L1 Lagrange point that saves nearly 59 metres per second of delta-v while keeping spacecraft in unbroken contact with Earth. The finding arrives as NASA prepares crewed lunar missions and quietly asks a larger question: how many other well-mapped roads in space are hiding better roads beneath them?
- NASA's Artemis II crew lost radio contact with Earth for 40 minutes in April when the Moon's own mass blocked their signal — a geometric inevitability that the new route eliminates entirely.
- Every metre per second of delta-v saved translates into exponentially less propellant at launch, meaning 58.80 m/s is not a rounding error but a meaningful reduction in mission cost and mass.
- The breakthrough required simulating 30 million trajectories using a technique called the theory of functional connections — ten times the scale of prior searches — revealing solutions that conventional optimization had missed for decades.
- The counterintuitive key was approaching lunar orbit from the Moon-facing side rather than the Earth-facing side, letting the natural gravitational architecture of the Earth-Moon system do the work.
- The method's limits are real — solar gravity and other perturbations are not yet modeled — but researchers believe the same computational approach could uncover hidden efficiencies across Mars, asteroid, and outer-planet missions.
A research team spanning Brazil and Portugal has charted a new path to the Moon that achieves something the space industry believed it had already optimized: it costs less fuel and never loses sight of Earth. The discovery came from a systematic search through 30 million possible trajectories — a computational scale made possible by a technique called the theory of functional connections — and it revealed what decades of conventional optimization had overlooked.
The route saves roughly 58.80 metres per second of delta-v, the measure of velocity change needed to complete a journey. In spaceflight, where every metre per second demands exponentially more propellant at launch, this is a meaningful gain. The savings come from a counterintuitive move: rather than approaching lunar orbit from the Earth-facing side, the spacecraft enters from the Moon-facing side, where the gravitational structure of the system offers more natural assistance.
The path works by routing spacecraft through L1, the gravitational balance point between Earth and Moon where a vehicle can hover indefinitely. This intermediate stop directly addresses a problem NASA's Artemis II crew encountered in April — a 40-minute communications blackout as their Orion capsule passed behind the Moon. The blackout was pure geometry, not malfunction, but for crewed missions it changes the calculus around emergencies, navigation, and abort decisions. The new trajectory eliminates it by ensuring the spacecraft always has a clear line of sight to Earth.
The model currently accounts only for Earth and Moon; adding the Sun's gravity could reveal even more efficient routes but would constrain launch windows — a familiar trade-off between precision and flexibility. The broader ambition, however, reaches past the Moon. Lead researcher Allan Kardec de Almeida Júnior hopes the method itself gets adopted widely. If systematic large-scale trajectory searches can find hidden savings on the lunar run, mission architectures built around Mars, asteroids, and the outer planets may be sitting on similar undiscovered efficiencies, waiting for someone to look.
A team of researchers working across Brazil and Portugal has mapped a new path to the Moon that does something the space industry thought it had already solved: it saves fuel while keeping spacecraft in constant contact with Earth. The discovery emerged from a systematic search through 30 million possible trajectories, a computational feat that revealed what conventional optimization had missed for decades.
The route, published in the journal Astrodynamics by a group led by Allan Kardec de Almeida Júnior at the University of Coimbra, cuts roughly 58.80 metres per second from the delta-v budget—the total change in velocity needed to complete the journey. In the language of spaceflight, this sounds like a rounding error. It is not. Every metre per second of delta-v translates into exponentially more propellant mass at launch, which is why mission planners spend years hunting for fractional improvements. The savings emerge from a counterintuitive insight: the cheapest way into lunar orbit is not the path closest to Earth, but the one that approaches from the Moon's side instead.
The trajectory works by routing spacecraft through the L1 Lagrange point, a gravitational balance point between Earth and the Moon where a spacecraft can hover indefinitely. This intermediate stop solves a problem that NASA's Artemis II crew encountered in April, when their Orion capsule passed behind the Moon for roughly 40 minutes and lost radio contact with Earth. The blackout was not a malfunction—it was geometry. The Moon itself blocked the signal. For crewed missions, any minutes out of contact change the calculus of medical emergencies, navigation decisions, and abort procedures. The new trajectory eliminates this entirely by keeping the spacecraft positioned where it can always see Earth.
What makes this discovery possible is a computational technique called the theory of functional connections, which dramatically reduces the cost of modelling complex orbital dynamics. Earlier studies simulated around 280,000 trajectories. This team simulated 30 million. The sheer breadth of the search revealed solutions that local optimization methods would never find—a hidden branch on a well-mapped highway.
The journey itself splits into two segments. The first carries the spacecraft from a 167-kilometre Earth parking orbit onto a stable manifold leading to L1. The second departs L1 and transitions into lunar orbit. The critical choice lies in that second segment: entering the lunar-orbit pathway from the Moon-facing side rather than the Earth-facing side, where the gravitational structure of the system offers more free assistance. This is how the savings accumulate—not from a single clever maneuver, but from riding the natural contours of space itself.
The model has limits. It accounts only for Earth and the Moon. Real spacecraft also feel the Sun's pull, radiation pressure, and perturbations from other bodies. Adding the Sun's gravity could reveal even more efficient routes, but it would tie any given trajectory to a specific launch date, narrowing the launch window. This is a familiar trade-off in trajectory design: precision versus flexibility.
The deeper significance lies not in a single number but in what it reveals about the lunar transportation problem. Half a century after Apollo 11, the cheapest known way to reach the Moon apparently was not the cheapest way at all. For a cislunar economy that will see dozens of crewed and uncrewed flights across the next decade, the implications extend far beyond this one paper. Almeida's hope is that the method itself—systematic search through massive solution spaces—gets adopted more widely. If this approach generalizes to Mars transfers, asteroid rendezvous, and outer-planet flybys, mission architectures built on conventional optimization may all be sitting on similar hidden savings, waiting for someone to look.
Notable Quotes
The orbit we propose is a solution that maintains uninterrupted communication.— Vitor Martins de Oliveira, postdoctoral researcher at University of São Paulo, referencing Artemis II
The Hearth Conversation Another angle on the story
Why does saving 58 metres per second matter so much? That sounds like nothing.
In rocketry, delta-v translates exponentially into propellant mass at launch. A small saving in velocity becomes a large saving in weight you have to lift off the ground. Over decades of missions, those savings compound into real money and real capability.
So this is just a better calculation of a path that already existed?
Not quite. The path existed in the mathematical space of all possible trajectories, but no one had looked there systematically. Conventional wisdom said the cheapest entry to lunar orbit came from the Earth-facing side. This team showed the Moon-facing side was actually better. It was hidden in plain sight.
And the communication problem—that's a separate benefit, or does it come from the same route?
It comes from the same route. By parking at L1 as an intermediate waypoint, the spacecraft never slips behind the Moon. It's always in line-of-sight with Earth. That solves two problems at once: fuel efficiency and continuous contact.
Why hasn't anyone found this before? The Moon isn't new.
Because finding it required simulating 30 million trajectories. Earlier studies could only handle about 280,000. The computational technique they used—functional connections—made the massive search feasible. You need both the math and the computing power.
Does this change how we'll actually fly to the Moon?
Not immediately. The model doesn't account for the Sun's gravity, radiation pressure, and other real-world forces. Adding those would make it more accurate but would tie trajectories to specific launch dates. But the method itself—systematic search through huge solution spaces—could reshape how we plan missions to Mars, asteroids, and beyond.
So this is really about a new way of thinking about trajectory design?
Exactly. The lunar problem was treated as solved. This shows it wasn't. If that's true for the Moon, it's probably true for everywhere else we want to go.