🔬 physics

Characterization of Lunar Escape Solution Space in the Bi-Circular Restricted Four-Body Problem

This paper develops a lobe-based framework utilizing Lagrangian Coherent Structures and Finite-Time Lyapunov Exponents to characterize the time-dependent L1-escape solution space in the Bi-Circular Restricted Four-Body Problem, offering a systematic tool for designing single-impulse lunar departure trajectories from low lunar orbits.

Original authors: Takayuki Shihara, Kathleen C. Howell, Yasuhiro Kawakatsu

Published 2026-09-24

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Takayuki Shihara, Kathleen C. Howell, Yasuhiro Kawakatsu

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the vast theater of space travel, the Moon is no longer just a distant destination; it is becoming a gateway. As humanity looks toward deeper space, the Moon serves as a critical staging point, a place where spacecraft must park, refuel, and then launch again on long journeys to Mars or beyond. To leave the Moon's neighborhood, a spacecraft cannot simply fire its engines and fly straight out. The gravitational pull of the Earth and the Sun creates a complex, shifting landscape of forces. In this environment, the path to freedom is not a straight line but a delicate curve that must thread through a narrow opening in the gravitational field. For decades, scientists have used a simplified model of the Earth-Moon system to map these paths, treating the Sun's influence as negligible. However, for precise mission planning, that simplification is no longer enough. The Sun's gravity is a constant, time-varying force that reshapes the available routes, turning a static map into a dynamic, moving target.

A team of researchers from the University of Tokyo, Purdue University, and the Japan Aerospace Exploration Agency has developed a new way to navigate this shifting terrain. They focused on the specific challenge of escaping the Moon's vicinity through the "L1 neck," a narrow gateway that connects the lunar region to the space between the Earth and the Moon. In a simplified model where the Sun is ignored, the paths to escape are fixed and unchanging, like roads on a static map. But when the Sun is included, the map changes every moment. The researchers realized that trying to find a single, perfect path in this moving environment was less useful than understanding the entire "solution space"—the full collection of all possible starting points that could lead to a successful escape. To do this, they created a framework that treats the escape routes not as isolated lines, but as a connected hierarchy of shapes, or "lobes," that evolve over time.

The core of their discovery relies on a method borrowed from fluid dynamics, used to track how water or air moves in complex currents. Instead of looking for fixed roads, the team looked for invisible barriers that separate water that stays in one place from water that flows away. In the context of space, these barriers separate spacecraft that remain trapped near the Moon from those that successfully escape. By analyzing the stretching and folding of trajectories over a short period, the researchers identified a specific region on a map of the Moon's vicinity called the "escape lobe." This lobe represents the set of all possible positions and speeds a spacecraft can have at its closest approach to the Moon (its perilune) that will allow it to slip through the L1 neck and escape without completing a full orbit around the Moon.

What makes this approach powerful is how it handles the Sun's influence. The researchers found that the size and shape of this escape lobe are not constant; they breathe and change depending on two main factors: the energy level of the spacecraft and the position of the Sun relative to the Earth-Moon line. When the Sun is in certain positions, the escape lobe expands, offering a wider range of starting conditions for a successful departure. In other positions, the lobe shrinks, making escape much harder or even impossible for certain energy levels. Crucially, the study showed that even if a spacecraft starts with too little energy to escape in a simplified model, the changing gravitational pull of the Sun can temporarily lower the energy barrier, allowing escape to occur at times when it would otherwise be forbidden. This means that the window for leaving the Moon is not just about how fast the ship is going, but also about exactly when it attempts the maneuver.

The researchers did not stop at the immediate escape. They used the escape lobe as a starting point and traced paths backward in time to find the "perilune lobes." These are the sets of starting conditions that lead to an escape after the spacecraft has circled the Moon one time, two times, or up to six times. This created a connected chain of possibilities, linking the final moment of escape back to earlier orbits. By visualizing these lobes in three dimensions, the team could see how the available paths shift and deform as the Sun moves. They discovered that the structure of these paths repeats in a predictable cycle, roughly every half of the Sun's orbit around the Earth-Moon system. This periodic behavior means that mission planners can predict with high confidence when the "doors" to escape will be wide open and when they will be closed.

To test the practical value of this framework, the team applied it to a realistic scenario: a spacecraft in a low lunar orbit, just 100 kilometers above the surface. They calculated the precise speed change required to jump from this stable orbit onto one of the escape paths. The results showed that the success of such a maneuver depends heavily on the timing of the departure. For a specific energy level, there are distinct windows of time, defined by the Sun's angle, where a single engine burn can send the spacecraft on a direct escape path. In other windows, the same engine burn might only allow the spacecraft to circle the Moon a few times before escaping, or it might fail to escape at all. The study demonstrated that by choosing the correct departure time, a mission can access a much wider variety of escape trajectories, including those that require less fuel or offer different flight times.

This work provides a new, systematic tool for designing lunar departure missions. Instead of guessing or running thousands of random simulations to find a single working path, engineers can now look at the "lobe" map to see the entire landscape of possibilities at a glance. They can identify the optimal times to leave the Moon to maximize their options and minimize fuel usage. The framework is particularly useful for preliminary mission design, where the goal is to understand the broad range of feasible solutions before committing to a specific trajectory. While the current study focused on a two-dimensional model of the solar system, the authors suggest that this lobe-based approach can be extended to more complex, three-dimensional scenarios and to escapes through other gateways, such as the L2 neck on the far side of the Moon. By turning a chaotic, time-dependent problem into a structured, visual map, this research offers a clearer path for the next generation of explorers leaving the lunar neighborhood.

Technical Summary: Characterization of Lunar Escape Solution Space in the Bi-Circular Restricted Four-Body Problem

Problem Statement
The design of lunar escape trajectories is a critical component of cislunar mission planning, serving as a staging point for deep-space missions. While the Circular Restricted Three-Body Problem (CR3BP) provides a foundational framework for analyzing transport structures via invariant manifolds and zero-velocity curves, it neglects solar gravity. In the Earth-Moon system, solar perturbations introduce explicit time dependence, rendering the Jacobi constant non-conserved and causing the time-invariant invariant manifolds of the CR3BP to dissolve. Consequently, the CR3BP's geometric transport structures are insufficient for characterizing escape in the more realistic Bi-Circular Restricted Four-Body Problem (BCR4BP). The challenge lies in identifying a systematic framework to describe the escape solution space in this non-autonomous environment, specifically for trajectories departing through the $L_1$ neck region.

Methodology
The authors develop a lobe-based framework to characterize the lunar escape solution space within the BCR4BP. The methodology integrates three primary components:

Dynamical Models: The study utilizes the CR3BP for baseline comparison and the BCR4BP, which includes the Sun as a fourth body moving on a circular orbit around the Earth-Moon barycenter. The BCR4BP is treated as a non-autonomous system where the Jacobi constant varies with time.
Lagrangian Coherent Structures (LCS) and FTLE: To identify finite-time transport barriers in the absence of time-invariant manifolds, the authors employ the Finite-Time Lyapunov Exponent (FTLE) field. Ridges in the FTLE field correspond to Lagrangian Coherent Structures (LCS), which act as separatrices between transit and non-transit orbits over a finite integration time ( $T=4.0$ , approx. 17.4 days).
Periapsis Poincaré Maps: The analysis is conducted on a Moon-centered polar coordinate section $(r, \phi)$ at perilune passages ( $\dot{r}=0, \ddot{r}>0$ ). This reduces the four-dimensional planar state space to a two-dimensional map.

The framework proceeds in two stages:

Identification of the Escape Lobe: The set of perilune states that escape through the $L_1$ neck without completing a single lunar revolution is identified as the "escape lobe." This is extracted by locating the closed ridge of high FTLE values on the periapsis section.
Construction of Perilune Lobes: Initial conditions that escape after multiple lunar revolutions are obtained by backward-time propagation from the escape lobe. The perilune states encountered along these backward trajectories form a hierarchy of "perilune lobes," indexed by the number of revolutions ( $k$ ) remaining before escape.

To address the non-autonomous nature of the system, the authors propose two representations:

Standard Representation: Parameterizes the solution space by the initial Jacobi constant ( $C_0$ ) and initial solar phase angle ( $\theta_{S0}$ ) at the moment of escape lobe crossing.
Alternative Representation: Re-indexes the perilune lobes by the instantaneous solar phase angle ( $\theta_S$ ) at the specific departure perilune passage, allowing for direct retrieval of solutions for a specified departure time.

Key Results

Escape Lobe Characteristics: The escape lobe is identified as a closed region bounded by LCS. Its area exhibits quasi-sinusoidal, $\pi$ -periodic variations with respect to the initial solar phase angle ( $\theta_{S0}$ ), with maxima near $100^\circ$ and $280^\circ$ and minima near $10^\circ$ and $190^\circ$ .
Parameter Dependence:
- Jacobi Constant ( $C_0$ ): As $C_0$ decreases (energy increases), the escape lobe area expands. Notably, in the BCR4BP, escape solutions exist even when $C_0 \ge C_1$ (where the $L_1$ neck is geometrically closed in the CR3BP) because the time-varying Jacobi constant can drop below $C_1$ during the trajectory.
- Solar Phase Angle ( $\theta_{S0}$ ): The shape and size of the lobe are strongly dependent on $\theta_{S0}$ . The variation is driven by the rate of change of the Jacobi constant ( $\dot{C}$ ), which is determined by the dot product of velocity and solar perturbation acceleration. When $\dot{C} < 0$ , the $L_1$ neck widens, enlarging the escape lobe.
Perilune Lobes: Backward propagation reveals a connected sequence of perilune lobes. The size of these lobes decreases monotonically as the number of revolutions increases. The boundaries of these lobes act as finite-time transport barriers, analogous to invariant manifolds in the CR3BP.
Low Lunar Orbit (LLO) Application: The framework is applied to single-impulse escape from a 100 km altitude LLO. By intersecting the "LLO cylinder" (representing the circular orbit extended in Jacobi constant space) with the lobe structures, the authors identify feasible transfer points.
- The intersection pattern is highly sensitive to $C_0$ and $\theta_{S0}$ . For instance, at $\theta_{S0} = 10^\circ$ , no single-impulse escape solutions were found for the tested parameters, whereas at $\theta_{S0} = 100^\circ$ , multiple solutions (direct and multi-revolution) were accessible.
- The alternative representation successfully demonstrated how to slice the solution space by a specific departure time ( $\theta_S$ ), directly yielding the accessible perilune states and required $\Delta V$ for that specific epoch.

Significance and Claims
The paper claims to provide a "lobe-based approach" for escape design in non-autonomous multi-body environments. The primary significance lies in offering a practical tool for the systematic search of initial conditions in preliminary lunar-departure design. By characterizing the escape solution space not as isolated trajectories but as a connected hierarchy of lobe structures, the framework bridges the gap between the geometric intuition of the CR3BP and the time-dependent reality of the BCR4BP.

The authors assert that this methodology enables mission designers to:

Systematically select candidate trajectories based on mission constraints (e.g., time of flight, $\Delta V$ ).
Identify favorable departure geometries (specifically solar phase angles) that maximize the accessible escape solution space.
Directly retrieve initial conditions for a specified departure time using the alternative representation.

The study concludes that while the framework is demonstrated for the $L_1$ neck and up to six revolutions, it is extensible to higher revolution counts and, in principle, to the three-dimensional BCR4BP and $L_2$ -side escape, though these remain future directions. The work does not claim to solve the full end-to-end Earth-return trajectory but focuses on the critical preliminary design phase of identifying dynamically admissible lunar-departure candidates.

More like this