The idea behind the LEGA is deceptively simple, yet it had never been flown before. In conventional interplanetary trajectory design, a gravity assist changes only the direction of a spacecraft’s velocity relative to a flyby body, not its magnitude. Mission analysts at ESA’s European Space Operations Centre in Darmstadt realised that instead of treating the Moon as a nuisance perturbation to be avoided during an Earth flyby, they could deliberately target it. By passing close to the Moon first, the spacecraft’s velocity relative to the Earth is altered in magnitude as well as direction, because the Moon’s own motion around the Earth is added to the spacecraft’s trajectory during the lunar encounter. For JUICE, the effect was a sharp increase in the Earth-relative escape velocity: the incoming value of roughly 3 kilometres per second was raised by about 300 metres per second during the Moon flyby, stabilising around 3.3 kilometres per second, which then remained constant until escape.

That 300 metres per second may sound modest, but its value is enormous. Had the spacecraft needed to generate the same velocity increase with its own engines, the optimal deep-space manoeuvre, applied far from Earth at aphelion, would have required around 150 metres per second of Delta-V. That translates into more than 100 kilograms of satellite mass that can instead be used for other purposes. JUICE needed every gram it could save: a direct transfer to Jupiter was beyond the performance of the Ariane 5 launcher that lifted the spacecraft from Kourou, French Guiana, on 14 April 2023, and beyond the capacity of its onboard propellant. The mission therefore depends on a chain of gravity assists, and the LEGA was selected into the baseline of the interplanetary cruise precisely because, on some launch days, it beats a standard Earth gravity assist in total Delta-V cost.

The choice came with a price, and the paper is candid about it. Gravity assists are non-linear events that rapidly amplify any dispersion in the spacecraft’s trajectory: a small navigation or guidance error before the flyby balloons into a large correction afterwards. In the LEGA case, the roughly 24 hours between the Moon flyby and the Earth flyby is far too short to implement a trajectory correction manoeuvre in between, so the initial dispersion is amplified twice, demanding a much larger cleanup than a standard Earth flyby would. Pre-launch analyses estimated the stochastic Delta-V cost at 75 metres per second, compared with a typical 15 metres per second for a standalone Earth gravity assist, and this difference partially eats into the deterministic savings. Safety was another consideration: the Moon closest approach was set as low as possible, at 750 kilometres from the lunar surface in the reference design, with the constraint of guaranteeing no impact, while the Earth pericentre altitude was governed by the need to encounter Venus later. Analyses of worst-case dispersions showed expected perigee altitude deviations of 200 kilometres at the 99 percent level, well above the atmosphere, so the manoeuvre was judged safe.

The navigation campaign itself unfolded like a slow-motion drama in the Moon’s B plane, the two-dimensional target plane used by navigators to plan planetary swing-bys. Trajectory correction slots were scheduled at four, two and one week before the Moon flyby, with a cleanup manoeuvre one week after the Earth flyby and a second planned near the subsequent perihelion on 1 January 2025. At the data cut-off for the first slot, on 16 July 2024, the orbit determination showed the spacecraft’s impact point significantly displaced from the target, well outside the three-sigma uncertainty ellipse. Correcting it early cost a Delta-V of just 3.8 centimetres per second; leaving it to be fixed after the LEGA would have demanded more than 160 metres per second. The manoeuvre was commanded, uplinked and executed flawlessly a few days later. By the two-week and one-week checkpoints, on 31 July and 6 August, the error ellipses had grown to or beyond the remaining dispersion, and the post-LEGA correction penalty had fallen below 10 metres per second, so no further manoeuvres were executed before the flybys.

As the spacecraft closed in on the Moon, tracking was continuously increased, and the successive estimates of the impact point remained stable while the error ellipses shrank, each nested inside the previous one. The reconstructed trajectory after the flyby showed an impact point just 2 kilometres from the optimum target, with a tiny error ellipse thanks to the strong information the bent trajectory provided about the spacecraft’s position relative to the Moon. The reconstructed parameters were strikingly precise: Moon closest approach on 19 August 2024 at 21:14:54.681 UTC with a three-sigma uncertainty of 0.001 seconds, at a distance of 2489.519 kilometres from the Moon’s centre, or 752 kilometres above its surface; Earth closest approach the following day at 21:56:13.637 UTC, at 13,210.572 kilometres from the Earth’s centre, or 6839 kilometres above the surface. The 2-kilometre dispersion was corrected with two post-LEGA manoeuvres, one on 27 August 2024 at 1.60 metres per second and one near perihelion on 1 January 2025 at 1.78 metres per second, both executed successfully, bringing the total stochastic Delta-V to 3.42 metres per second.

From the operations side, the spacecraft performed superbly through a genuinely punishing sequence. The Earth flyby occurred while the Moon was full, with JUICE arriving from the dark side, so the spacecraft endured a near-simultaneous lunar eclipse and ground station occultation shortly before the Moon closest approach. The batteries had been fully charged four weeks in advance to cope with the 32-minute Moon eclipse, during which they discharged by around 6 percent. In a first for the mission, four solar array power regulators were observed in maximum power point tracking mode during the eclipse penumbra, a mode never before seen in flight. All star tracker optical heads were blinded by the Earth for around two hours, though never all three at once, and the attitude control system continued nominally throughout without ever having to rely solely on the internal gyroscopic measurement unit.

Not everything was perfectly smooth, and the paper documents the anomalies with the same care. Flight dynamics noticed a small torque and Delta-V beginning about five minutes after the Moon closest approach and lasting around 20 minutes, coinciding with sunlight reflected off the Moon illuminating the +X face of the spacecraft, a face normally kept in the dark. The effect is suspected to be spacecraft outgassing and was confirmed by several instrument measurements. The laser altimeter GALA was the only payload affected by an anomaly, suffering an unexplained instrument reset that prevented it from measuring at the Moon closest approach. Communications also demanded careful handling: the spacecraft switched from its medium-gain to its low-gain antenna three days around the Earth flyby, the transmitter was switched off near closest approach to comply with ITU regulations on transmitting at full power below 45,000 kilometres, and unfavourable low-gain antenna visibility meant no communication was attempted, producing a total blackout of more than eight hours.

Science operations ran remarkably smoothly, thanks in part to a full rehearsal of the LEGA operations in early July 2024. All instruments operated during the flyby, with observation slots at the Moon from one hour before to one hour after closest approach and at the Earth from four hours before to three days after. The navigation camera captured 15 images, six of the Moon and nine of the Earth, while the two JUICE Monitoring Cameras obtained 140 and 144 images respectively, yielding striking portraits of both worlds. The spacecraft was kept with its instrument optical bench shielded behind the high-gain antenna in a Sun-pointing attitude, with a series of precisely timed rolls optimising the viewing geometry for the payloads. As a bonus, a dedicated farewell off-pointing slew on 9 September 2024 allowed several payloads to observe the Earth-Moon system along the outbound arc, within the 45-minute off-pointing limit allowed at Earth distances.

The overall verdict from the team is unambiguous: the LEGA was an outstanding operational and technical success, demonstrating ESA’s ability to execute complex multi-body gravity assists with exceptional precision and preserving almost the entire stochastic Delta-V budget. That preserved propellant directly enhances the mission’s scientific potential at Jupiter, where JUICE will conduct 35 flybys of Ganymede, Europa and Callisto before becoming the first spacecraft to orbit a moon of the outer Solar System, ultimately settling into a circular orbit around Ganymede. The mission was precisely on track for its Venus flyby in August 2025 and its later Earth flybys en route to the Jovian system, with the LEGA now validated as an enabling technique for the complex interplanetary trajectories of the future.

Subject of Research: The trajectory design, navigation and spacecraft operations of the first Lunar–Earth gravity assist performed by ESA's JUICE mission in August 2024.

Article Title: The JUICE Lunar–Earth gravity assist from trajectory design, navigation and spacecraft operations perspective

Article References: Dietz, A., Boutonnet, A., & Budnik, F. (2026). The JUICE Lunar–Earth gravity assist from trajectory design, navigation and spacecraft operations perspective. Annales Geophysicae, 44(2), 1131-1142. https://doi.org/10.5194/angeo-44-1131-2026

Image Credits: AI Generated

DOI: 10.5194/angeo-44-1131-2026

Keywords: JUICE, ESA, gravity assist, Lunar-Earth flyby, trajectory design, spacecraft navigation, trajectory correction manoeuvre, Jupiter, Venus flyby, interplanetary cruise, flight dynamics, propellant budget