Information-Based Trajectory Planning for Spacecraft-to-Spacecraft Tracking and Navigation in Cislunar Space

Deep space is a place where the familiar rules of navigation begin to fray. For decades, spacecraft traveling near Earth have relied on a vast network of ground-based antennas to tell them where they are and where they are going. These antennas, scattered across the planet, listen for radio signals from the ships and calculate their positions with great precision. However, as humanity looks to expand its presence into the space between the Earth and the Moon, this reliance on Earth becomes a liability. The distances are too great for real-time communication, and the sheer volume of the region makes it impossible for a few antennas on the ground to watch everything at once. Furthermore, the gravitational pull of the Earth, the Moon, and the Sun creates a complex, shifting environment where a ship's path is never a simple straight line or a perfect circle. To operate safely and effectively in this "cislunar" space, spacecraft will need to become far more independent, capable of finding their own way without constant help from home.

One promising solution is for spacecraft to help each other. Instead of waiting for a signal from Earth, an observer ship can track a target ship by measuring the angle and direction to it, much like a sailor spotting a lighthouse. This is called spacecraft-to-spacecraft tracking. The challenge lies in the geometry of the situation. If two ships are too close together, or if they move in a way that looks too similar, the observer cannot easily tell where the target truly is. The information gained from the measurements is muddy. To solve this, the ships must move in a way that creates the clearest possible picture of their relative positions. This requires a delicate balance: the ships must maneuver to get the best view, but every time a ship fires its engine to change course, it uses fuel and introduces new uncertainties about its own exact location. The question for engineers is how to move the ships to learn the most about their surroundings while spending the least amount of fuel.

In a recent study, researchers at the University of Colorado Boulder and the University of Texas at Austin developed a new method to solve this problem. They created a computer algorithm that plans the flight paths of low-thrust spacecraft—ships that use gentle, continuous pushes from their engines rather than powerful, short bursts. The goal of their method is to design a trajectory that maximizes the information the observer gathers about the target, while keeping the fuel cost reasonable. The researchers treated the problem as a trade-off. They built a mathematical model that calculates how much "information" a specific flight path would generate. This information is not just about seeing the target; it is about understanding the subtle differences in how the two ships react to the complex gravitational forces of the Earth-Moon system. By analyzing these differences, the observer can figure out the absolute position of both ships with much greater accuracy than if they just drifted along.

The team tested their method using a simulation of two spacecraft orbiting in the Earth-Moon system. They placed both ships in a specific type of path called a Distant Retrograde Orbit, which is a stable loop that goes around the Moon in the opposite direction of its rotation. In their simulation, one ship acted as the observer, equipped with an optical sensor to watch the other ship, which acted as the target. The researchers ran the simulation many times, changing the weight they gave to "information" versus "fuel." When they told the computer to prioritize information heavily, the observer ship did something counterintuitive. Instead of staying close to the target to get a better look, it moved further away. This happened because, in the complex gravitational environment of the Earth-Moon system, moving apart creates a greater difference in how the two ships are pulled by gravity. This difference, or "dynamical diversity," helps the observer distinguish the target's true path from its own, leading to a much clearer picture of where both ships are.

The results of the simulation were striking. When the researchers included this information-based planning in the flight path, the expected error in knowing where the ships were dropped by nearly a factor of ten compared to a path that only tried to save fuel. In other words, the ships could know their location with ten times more precision. This improvement was most pronounced when the observer ship started with a poor understanding of its own position. In those cases, the algorithm correctly decided to fly a path that emphasized the differences in motion between the two ships, even if it meant using more fuel. However, if the observer already knew its position very precisely, the algorithm shifted its strategy. It then chose a path that brought the observer closer to the target, maximizing the geometric detail of the view. The method proved flexible enough to adapt to the starting conditions of the mission.

The researchers achieved this by using a sophisticated optimization technique that breaks the continuous flow of time into small steps, allowing the computer to solve the complex problem piece by piece. They also developed a way to estimate the information gain without needing to run thousands of random simulations, which would be too slow for real-time planning. Instead, they used a mathematical approximation that captures the essential behavior of the system. This allowed them to find the best path quickly, even on a standard laptop computer. The study confirms that for future missions in cislunar space, such as those planned for the Lunar Gateway, spacecraft will need to be active participants in their own navigation. They cannot simply drift; they must maneuver with purpose, using their engines to sharpen their view of the universe around them.

This work does not claim to have solved every problem of deep space navigation. The study was conducted in a simulation using a simplified model of gravity, and real missions will have to contend with the full complexity of the solar system, including the effects of sunlight and the Moon's irregular shape. The researchers acknowledge that future work will need to incorporate these factors and test the method with multiple ships and different types of sensors. However, the core finding is robust: by treating the act of navigation as a way to gather information, rather than just a way to get from point A to point B, spacecraft can achieve a level of self-reliance that was previously out of reach. The ability to trade a small amount of fuel for a massive gain in situational awareness could be the key to unlocking the next era of lunar exploration, allowing ships to navigate the vast, silent space between worlds with confidence.