🔭 astrophysics

In-orbit Demonstration of X-ray Pulsar Navigation with NinjaSat

This study demonstrates the feasibility of X-ray pulsar navigation using the NinjaSat CubeSat by verifying stable timing performance and achieving position errors between 27 and 370 km through the SEPO method, while experimentally confirming that navigation accuracy depends on the seasonal geometry between the orbital plane and the pulsar direction.

Original authors: Naoyuki Ota, Takuya Takahashi, Toru Tamagawa, Tomoshi Takeda, Teruaki Enoto, Takao Kitaguchi, Wataru Iwakiri, Yo Kato, Masaki Numazawa, Tatehiro Mihara, Hiromitsu Takahashi, Chin-Ping Hu, Yuanhui Zhou

Published 2026-02-19

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Naoyuki Ota, Takuya Takahashi, Toru Tamagawa, Tomoshi Takeda, Teruaki Enoto, Takao Kitaguchi, Wataru Iwakiri, Yo Kato, Masaki Numazawa, Tatehiro Mihara, Hiromitsu Takahashi, Chin-Ping Hu, Yuanhui Zhou, Keisuke Uchiyama, Yuto Yoshida, Syoki Hayashi, Arata Jujo, Sota Watanabe, Amira Aoyama, Satoko Iwata, Kaede Yamasaki, Soma Tsuchiya, Yosuke Nakano, Takayuki Kita, Mayu Ichibakase, Hiroki Sato, Hirokazu Odaka, Tsubasa Tamba, Kentaro Taniguchi

Original paper licensed under CC BY 4.0 (http://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

Imagine you are driving a car in a dense fog, far away from any road signs, streetlights, or GPS satellites. You have no idea where you are, only that you are moving. How would you find your way?

In deep space, astronauts face this exact problem. Once a spacecraft leaves Earth's orbit, the GPS signals we use on our phones stop working. For decades, scientists have proposed a solution: X-ray Pulsar Navigation.

Think of pulsars as the "lighthouses of the universe." They are dead stars (neutron stars) that spin incredibly fast, beaming X-rays toward Earth like a cosmic strobe light. Because their spin is so stable, they act like the most precise clocks in the galaxy. If you can count the "ticks" of these cosmic clocks, you can figure out exactly where you are in space.

This paper describes a groundbreaking experiment where a tiny satellite, NinjaSat, successfully used this method to navigate. Here is the story of how they did it, explained simply.

1. The Tiny Detective: NinjaSat

Usually, to catch these faint X-ray signals from deep space, you need a massive, expensive telescope the size of a bus. But the team wanted to prove that a CubeSat—a satellite the size of a shoebox (about 11x24x34 cm)—could do the job.

NinjaSat is equipped with two special "eyes" called Gas Multiplier Counters (GMCs). Imagine these as highly sensitive rain gauges, but instead of rain, they catch individual X-ray photons. Even though they are small, they are sharp enough to catch the "ticks" of the Crab Pulsar, a famous cosmic lighthouse.

2. The Challenge: Keeping Perfect Time

To navigate, you don't just need to see the light; you need to know exactly when it arrives. If your watch is off by even a tiny fraction of a second, your calculation of where you are will be miles off.

The team had to build a super-precise timing system inside the shoebox satellite.

The Analogy: Imagine trying to time a sprinter with a stopwatch that has a slight delay. First, they calibrated their "stopwatch" (the satellite's internal clock) against the "official time" (GPS signals).
The Result: They proved their satellite could measure time with an accuracy of about 100 microseconds (that's 0.0001 seconds). To put that in perspective, light travels about 30 kilometers in that tiny slice of time. Their clock was accurate enough to keep them on track.

3. The Navigation Trick: The "SEPO" Method

How do you turn a blinking light into a map? The team used a clever mathematical trick called SEPO (Significance Enhancement of Pulse-profile with Orbit-dynamics).

The Analogy: Imagine you are in a dark room with a spinning fan that makes a "whoosh" sound every time a blade passes a microphone. You don't know where the fan is, but you know the sound is rhythmic.
- If you guess the fan's location wrong, the rhythm sounds messy and out of sync.
- If you guess the location right, the rhythm becomes perfectly clear and sharp.
The Process: The computer on NinjaSat tried thousands of different guesses about where the satellite was. For each guess, it calculated: "If we were here, would the pulsar's signal look sharp?"
- If the signal looked fuzzy, the computer said, "Wrong guess!"
- If the signal looked sharp and clear, the computer said, "Bingo! We are probably here."

This is like tuning a radio. You twist the dial until the static disappears and the music becomes clear. The "dial" in this case was the satellite's position.

4. The Twist: The Angle Matters

The most fascinating discovery in this paper is that the "tuning" works better at certain times of the year.

The Analogy: Imagine trying to hear a whisper from a friend standing on a hill.
- If you are walking in a circle around the hill, you can hear the whisper get louder and softer as you move. This helps you figure out exactly where you are.
- But if you are walking in a circle parallel to the hill, the distance to your friend never changes. You can't tell if you are moving forward or backward just by listening.

The team found that when the satellite's orbit was at a "sweet spot" angle relative to the Crab Pulsar, the navigation was incredibly accurate (within 27 to 60 kilometers). But when the angle was "flat" (like walking parallel to the hill), the satellite got confused about its position, and the error grew to hundreds of kilometers.

5. Why This Matters

This study is a huge leap forward for space exploration for three reasons:

Small is Beautiful: They proved you don't need a billion-dollar telescope to navigate deep space. A shoebox-sized satellite can do it. This means future missions could be cheaper and more numerous.
Self-Reliance: Deep space is lonely. GPS doesn't reach Mars or the moons of Jupiter. This proves that spacecraft can navigate themselves using the stars, without needing to call Earth for help.
The "Cosmic Compass": They showed that by understanding the geometry of the orbit (the angle), we can pick the best times to navigate, making the system much more reliable.

The Bottom Line

NinjaSat successfully used the "ticks" of a distant, spinning dead star to tell the tiny satellite where it was in Earth's orbit. It's like a tiny car driving through a foggy night, using the rhythmic flash of a distant lighthouse to know it's staying on the road.

This experiment is the first real-world proof that CubeSats can navigate themselves using the universe's own clock, paving the way for a future where small, cheap satellites can explore the deep cosmos on their own.

1. Problem Statement

Deep space exploration requires autonomous navigation systems because Global Navigation Satellite Systems (GNSS) like GPS do not provide coverage beyond Earth orbit. X-ray pulsar navigation (XNAV) is a proposed solution, utilizing the periodic X-ray signals from neutron stars (pulsars) as celestial beacons to estimate a spacecraft's position.

However, existing demonstrations of XNAV have largely relied on large-scale observatories (e.g., Insight-HXMT, NICER) with significant mass and power budgets. A critical gap exists in demonstrating the feasibility of XNAV on CubeSat-class platforms, which have severe constraints on size, mass, and power. Furthermore, while the "Significance Enhancement of Pulse-profile with Orbit-dynamics" (SEPO) method has been validated on large satellites, its performance on small satellites with limited effective areas and single-pulsar observation capabilities needed verification.

2. Methodology

The study utilized NinjaSat, a 6U CubeSat X-ray observatory, to demonstrate XNAV capabilities in Low Earth Orbit (LEO).

Instrumentation: NinjaSat is equipped with two Gas Multiplier Counters (GMCs). These are compact (1U each), sensitive to the 2–50 keV band, with an effective area of 16 cm² at 6 keV. The system includes a flight computer (FC), payload controller (PC), and a GPS receiver for time synchronization and ground-truth validation.
Timing Calibration:
- X-ray photon arrival times were recorded with a resolution of ~61 µs.
- Absolute timing was synchronized using GPS Pulse-Per-Second (PPS) signals.
- Timing accuracy was verified against the Crab Pulsar using the Jodrell Bank Observatory (JBO) radio ephemeris and cross-correlated with simultaneous observations by the NICER mission.
Navigation Algorithm (SEPO):
- The study employed the SEPO method, which treats orbital element estimation as a Bayesian optimization problem.
- Objective Function: The algorithm searches for orbital parameters that maximize the pulse profile significance ( $\chi^2$ ) of the Crab Pulsar.
- Optimization Parameters: The algorithm simultaneously optimized five orbital elements: atmospheric drag ( $B$ ), inclination ( $i$ ), mean motion ( $n$ ), orbital phase at start ( $\theta$ ), and right ascension of the ascending node ( $\Omega$ ).
- Propagation: The Simplified General Perturbations 4 (SGP4) model was used to propagate satellite positions based on the trial orbital elements.
Observation Strategy:
- Four distinct observation epochs were selected (April 2024, November 2024, January 2025, February 2025) to test the method under different geometric configurations between the satellite's orbital plane and the Crab Pulsar direction.
- Total exposure times ranged from ~80 ks to ~100 ks per epoch.

3. Key Contributions

First CubeSat XNAV Demonstration: This is the first experimental verification of pulsar navigation using a CubeSat-class X-ray observatory, proving that compact, low-power instruments can achieve viable navigation performance.
Validation of SEPO on Small Satellites: The study successfully adapted the SEPO method for a small satellite, utilizing Bayesian optimization to handle the limited effective area and single-pulsar constraints.
Geometric Dependence Analysis: The paper provides the first experimental evidence quantifying how the seasonal geometry between the orbital plane and the pulsar direction critically affects navigation accuracy.
High-Precision Timing: The study established that NinjaSat's timing system is stable within 100 µs, with a measured phase offset relative to the radio reference of $\sim$ 510 µs (consistent with known X-ray/radio lags) and a relative offset to NICER of $8 \pm 1$ µs.

4. Results

Timing Performance:
- The Crab Pulsar pulse profile was clearly detected.
- The absolute timing offset relative to the JBO radio ephemeris was consistent at $(-5.1 \pm 0.4) \times 10^2$ µs.
- Cross-correlation with NICER data showed a mean time offset of $8 \pm 1$ µs with a standard deviation of 21 µs, confirming high timing fidelity.
Navigation Accuracy:
- Line-of-Sight (LOS) Accuracy: The position component along the Crab Pulsar line of sight was consistently constrained within $\sim$ 40 km across all epochs.
- 3D Accuracy: The three-dimensional position error (Euclidean norm) varied significantly based on the observation epoch:
  - April 2024: $\sim$ 370 km (Unstable optimization due to large separation angle).
  - November 2024: $\sim$ 27 km (Highly stable).
  - January 2025: $\sim$ 49 km.
  - February 2025: $\sim$ 53 km.
Geometric Sensitivity:
- When the pulsar direction was nearly perpendicular to the orbital plane (e.g., April 2024), short-term orbital motion could not be constrained, leading to poor 3D accuracy.
- When the pulsar direction was at an optimal angle (approx. 45° to the orbital plane, as in Nov 2024 and Feb 2025), variations in all orbital elements were effectively projected onto the line of sight, yielding the most stable and accurate results.
- The SEPO-derived orbits remained stable over time, whereas orbits derived solely from initial NORAD TLEs degraded by tens of kilometers within a week.

5. Significance

This study demonstrates the feasibility of autonomous deep-space navigation using CubeSat-class hardware. It confirms that:

Compact Detectors Work: Small Gas Multiplier Counters can provide sufficient timing precision for XNAV.
Algorithmic Robustness: Bayesian optimization of orbital elements using a single pulsar is a viable strategy for small satellites, provided the observation geometry is favorable.
Operational Utility: Pulsar navigation can serve as a reliable backup or update mechanism for orbital elements when GPS or TLE data becomes unavailable or degraded, particularly in deep space where GNSS is absent.

The findings suggest that future missions should carefully select target pulsars based on the spacecraft's orbital dynamics to maximize navigation accuracy, and that CubeSat constellations could potentially form a distributed X-ray navigation network for deep space exploration.