Explainer | China is making rapid gains in space tech. Here’s how the military could use it | South China Morning Post

China's Space Military Ambitions: A Growing Challenge to US Dominance

China achieves historic space refueling milestone as Pentagon warns of rapidly expanding military space capabilities

July 7, 2025

China has achieved a groundbreaking milestone in space technology with the successful completion of the world's first in-orbit refueling mission at geostationary altitude, while simultaneously expanding its military space capabilities at an unprecedented pace that U.S. officials describe as a direct challenge to American space dominance.

Historic Space Refueling Achievement

In February 2025, China's Shijian-25 satellite successfully refueled the BeiDou G7 navigation satellite at an altitude of 30,000 kilometers, extending the satellite's operational life by eight years. This marks the first time any nation has conducted satellite refueling in geostationary orbit, a technological feat that dramatically reduces the cost of maintaining space infrastructure and represents what analysts call a new era of "space equality."

The mission builds on China's previous demonstration of satellite servicing capabilities, including the Shijian-21 satellite's successful relocation of a defunct BeiDou satellite to a "graveyard orbit" in 2022 using robotic arms and advanced maneuvering systems.

Rapid Military Space Expansion

According to the Pentagon's December 2024 China Military Power Report, China has dramatically expanded its space capabilities, growing from operating just 36 satellites in 2010 to over 1,000 satellites by 2024. More than 360 of these satellites are dedicated to intelligence, surveillance, and reconnaissance (ISR) missions, providing the People's Liberation Army (PLA) with unprecedented monitoring capabilities over U.S. and allied forces in the Pacific region.

Key developments include:

• Aggressive Satellite Maneuvering: In 2024, five Chinese satellites conducted what U.S. Space Force officials described as "dogfighting in space" – synchronized proximity operations in low-Earth orbit featuring multiple satellites maneuvering within kilometers of each other, practicing tactics for potential satellite-to-satellite operations.
• Advanced Communication Networks: China is developing two massive satellite constellations to rival SpaceX's Starlink. The Qianfan (Thousand Sails) constellation aims to deploy 14,000 satellites by 2030, while the Guowang (National Network) system targets 13,000 satellites to provide military-grade communications during conflicts.
• Enhanced Military Organization: China created an independent Aerospace Force in April 2024, directly under the Central Military Commission, signaling the elevated priority of space warfare capabilities.

BeiDou Navigation System: A GPS Alternative

China's BeiDou Navigation Satellite System, completed in 2020 with 60 satellites, now provides global positioning services that rival and in some areas exceed U.S. GPS capabilities. The Harvard Belfer Center notes that BeiDou offers superior accuracy in the Asia-Pacific region and unique features like two-way messaging that GPS lacks.

Military applications include:

• Independent guidance for Chinese missiles and precision munitions
• Enhanced accuracy for the PLA across all service branches
• Secure military communications through satellite messaging
• Support for "multi-domain precision warfare" strategies

Intelligence and Surveillance Capabilities

China operates multiple satellite families for military reconnaissance:

• Yaogan satellites: Equipped with optical reconnaissance, synthetic-aperture radar, and electronic intelligence sensors
• Gaofen ("High Resolution") satellites: Using optical, multispectral, radar, and radio frequency sensors for Earth observation
• Tongxin Jishu Shiyan (TJS) satellites: Providing early warning and signals intelligence for the PLA

The U.S. Space Force reports that these satellites are actively "stalking" American satellites, conducting unusual orbital maneuvers to closely approach and monitor U.S. space assets.

Dual-Use Space Capabilities

China's space program deliberately blurs the line between civilian and military applications. Technologies marketed for peaceful purposes often have clear military potential:

• Space debris removal: Satellites equipped with robotic arms and nets for cleaning space junk can also capture and destroy enemy satellites
• Reusable space planes: China's classified Shenlong ("Divine Dragon") spacecraft, similar to the U.S. X-37B, can deploy payloads, conduct surveillance, or potentially capture adversary satellites
• Anti-satellite weapons: China demonstrated direct-ascent anti-satellite capabilities in 2007 and continues developing ground-based lasers and jamming systems

Strategic Implications and Future Plans

China's October 2024 "National Space Science Medium and Long-Term Development Plan" outlines ambitious goals to become the world's preeminent space power by 2050. The three-phase plan includes:

1. 2025-2027: Solidifying technological foundations and unmanned exploration
2. 2028-2035: Landing taikonauts on the Moon and establishing a permanent lunar base
3. 2036-2050: Manned missions to Mars and other celestial bodies

The plan directly challenges U.S. space leadership and aims to position China at the "international forefront" of space-faring nations.

Advanced Technologies and AI Integration

The Pentagon's 2024 report highlights China's integration of artificial intelligence, quantum technology, and biotechnology into its military space strategy. Key developments include:

• Algorithmic warfare: AI-enabled capabilities for network-centric warfare by 2030
• Quantum communications: Plans to launch quantum communication satellites in 2025 for secure military communications
• Brain-computer interfaces: Research into neurocognitive warfare capabilities
• Space-based solar power: A demonstration satellite planned for 2025 to beam power back to Earth

International Concerns and Competition

The rapid pace of China's space militarization has prompted international concern. European think tanks criticize China's "lack of transparency," while U.S. officials warn that China is "defining the standards for the next generation of space infrastructure."

The development comes as the U.S. faces its own space challenges, with some assessments noting that GPS capabilities "are now substantially inferior to those of China's BeiDou" in several key areas.

China's space achievements also extend its influence through the "Space Silk Road," offering BeiDou services to Belt and Road Initiative countries and providing an alternative to Western space infrastructure for developing nations.

Looking Ahead

As China continues its aggressive space expansion, the competition for space dominance is intensifying. With breakthrough technologies like orbital refueling, advanced satellite constellations, and military space planes, China is rapidly closing the gap with the United States and potentially creating new asymmetric advantages in space-based warfare capabilities.

The success of missions like Shijian-25's refueling operation demonstrates that China's space ambitions are not merely aspirational but represent tangible capabilities that could reshape the strategic balance in space for decades to come.

---

Sources and Citations

1. South China Morning Post - "China is making rapid gains in space tech. Here's how the military could use it" -  Explainer | China is making rapid gains in space tech. Here’s how the military could use it | South China Morning Post
2. Foundation for Defense of Democracies - "Showcasing Advanced Space Capabilities, China Displays 'Dogfighting' Maneuvers in Low Earth Orbit" (March 21, 2025) - https://www.fdd.org/analysis/policy_briefs/2025/03/21/showcasing-advanced-space-capabilities-china-displays-dogfighting-maneuvers-in-low-earth-orbit/
3. Modern Diplomacy - "China's Strategic Ascent in Space: New Dynamics in 2025" (January 22, 2025) - https://moderndiplomacy.eu/2025/01/22/chinas-strategic-ascent-in-space-new-dynamics-in-2025/
4. HPC Wire - "DoD's 2024 China Report Highlights Plans for AI and Quantum in Military Use" (January 8, 2025) - https://www.hpcwire.com/2025/01/07/dods-2024-china-report-highlights-plans-for-ai-and-quantum-in-military-use/
5. Defense One - "China's plan to rule the heavens" (January 17, 2025) - https://www.defenseone.com/ideas/2025/01/china-space-science-dominance-moon/402294/
6. SpaceNews - "China launches Shijian-25 satellite to test on-orbit refueling and mission extension technologies" (January 6, 2025) - https://spacenews.com/china-launches-shijian-25-satellite-to-test-on-orbit-refueling-and-mission-extension-technologies/
7. SpaceNews - "Pentagon report highlights China's space advancements and AI-driven 'precision warfare'" (December 18, 2024) - https://spacenews.com/pentagon-report-highlights-chinas-space-advancements-and-ai-driven-precision-warfare/
8. Federation of American Scientists - "The 2024 DOD China Military Power Report" (December 18, 2024) - https://fas.org/publication/the-2024-dod-china-military-power-report/
9. The Nation Newspaper - "China Achieves Space Refueling Technology: A New Era of 'Space Equality' Dawns" (February 24, 2025) - https://thenationonlineng.net/china-achieves-space-refueling-technology-a-new-era-of-space-equality-dawns/
10. Andrew S. Erickson - "Latest 'Space Threat Fact Sheet' & Annex from Headquarters Space Force Intelligence" - https://www.andrewerickson.com/2025/04/latest-space-threat-fact-sheet-annex-from-headquarters-space-force-intelligence/
11. Light Reading - "China sets ambitious goals for its BeiDou satellites" (November 29, 2024) - https://www.lightreading.com/satellite/china-sets-ambitious-goals-for-its-beidou-satellites
12. International Defense Security & Technology - "China's Escalating Space Militarization: Assessing Destructive 'Counter-Space' Capabilities and Weapons Advancements" - https://idstch.com/space/chinas-escalating-space-militarization-assessing-destructive-counter-space-capabilities-and-weapons-advancements/
13. TS2 Space Technology - "Inside China's Space Empire: Satellites, Services, and the Secret Power of CNSA" - https://ts2.tech/en/inside-chinas-space-empire-satellites-services-and-the-secret-power-of-cnsa/
14. Wikipedia - "Chinese space program" (Updated 2 weeks ago) - https://en.wikipedia.org/wiki/Chinese_space_program
15. Wikipedia - "BeiDou" (Updated 4 days ago) - https://en.wikipedia.org/wiki/BeiDou

Autonomous Management Systems for Large-Scale Satellite Constellations

How AI Is Becoming the Ultimate Space Traffic Controller for Thousands of Satellites

As mega-constellations fill Earth's orbit, artificial intelligence steps in to prevent cosmic chaos—and it's working better than anyone expected

By Sarah Chen | Science Correspondent

Imagine trying to coordinate the flight paths of 50,000 aircraft, all traveling at 17,500 miles per hour, in a three-dimensional space with no air traffic controllers, no traffic lights, and no room for error. Welcome to the reality facing satellite operators today as Earth's orbit becomes increasingly crowded with mega-constellations that are revolutionizing everything from internet access to climate monitoring.

The numbers are staggering: more than 8,000 active satellites currently zip around our planet, with companies like SpaceX, Amazon, and OneWeb planning to launch tens of thousands more by 2030. Managing this celestial traffic jam has become one of the most complex logistical challenges ever attempted—and increasingly, we're turning to artificial intelligence to solve it.

The Space Highway Is Getting Crowded

"We're essentially building highways in space," explains Dr. Maria Rodriguez, a space systems engineer at MIT. "But unlike terrestrial highways, there are no speed limits, no lane markers, and if two vehicles collide, the debris can take out dozens of other 'cars' in a catastrophic chain reaction."

This nightmare scenario, known as Kessler Syndrome, could potentially render entire orbital zones unusable for decades. With satellites traveling at speeds where a paint fleck becomes a deadly projectile, precision timing isn't just important—it's existential.

The traditional approach of manually tracking each satellite simply doesn't scale. SpaceX's Starlink constellation alone includes over 5,000 satellites, each generating terabytes of data daily while constantly adjusting their positions to avoid space debris and each other. Managing this complexity manually would require thousands of operators working around the clock.

Enter the Robot Overlords (The Helpful Kind)

Instead, a diverse ecosystem of companies and research institutions are deploying AI systems that can think faster than any human operator. These digital air traffic controllers are proving remarkably effective at their cosmic juggling act—and the competition to build the best "space brain" is fierce.

The Commercial Space Race 2.0

Spire Global leads the pack with their Constellation Management Platform, which can simultaneously monitor and control over 100 satellites with just one human operator. The AI continuously analyzes orbital mechanics, predicts potential collisions, and automatically executes avoidance maneuvers—all in real-time.

"The AI can process scenarios that would take a human team days to analyze, and it does it in seconds," says Jennifer Park, Spire's director of mission operations. "It's like having a chess grandmaster who can think 50 moves ahead, except the chess board is three-dimensional and all the pieces are moving at hypersonic speeds."

Not to be outdone, Cognitive Space has developed CNTIENT.Optimize, an AI platform that processes over 50 terabytes of operational data daily. Their system doesn't just manage satellites—it predicts what Earth observation data will be most valuable and automatically prioritizes collection missions. Think of it as a crystal ball that tells satellites where to look before anyone even asks.

Meanwhile, aerospace giant Raytheon Technologies is revolutionizing ground operations with their AI-enabled systems that push software updates to satellite management platforms every two weeks—faster than most smartphone apps get updated. Their "superhuman eyes" AI can spot anomalies in satellite behavior 40% faster than human operators, often catching problems before they become critical failures.

a.i. solutions, a company that sounds like it came straight out of science fiction, has created FreeFlyer software that distributes complex orbital calculations across cloud computing networks. When SpaceX needs to calculate the trajectories for thousands of Starlink satellites, FreeFlyer can crunch the numbers 85% faster than traditional single-computer systems.

Academic Powerhouses Join the Fray

Universities aren't just watching from the sidelines—they're driving some of the most innovative research in the field.

Stanford University's Space Rendezvous Laboratory is pioneering precision formation flying algorithms that keep satellites positioned within meters of each other across vast orbital distances. Their work is enabling coordinated Earth observation missions that can create 3D models of natural disasters in real-time.

MIT's Space Systems Laboratory is tackling one of the field's biggest challenges: teaching satellites to fix themselves. Their research into autonomous on-orbit servicing could eliminate the need to abandon expensive satellites when components fail. Imagine if your car could drive itself to the mechanic and perform its own repairs—that's essentially what MIT is trying to achieve in space.

Carnegie Mellon University is developing swarm intelligence algorithms inspired by how bees coordinate their hives. Their research could enable hundreds of tiny CubeSats to work together as a single, massive sensor array, revolutionizing everything from weather prediction to asteroid detection.

The University of Colorado Boulder is home to the Laboratory for Atmospheric and Space Physics, where researchers are creating AI systems that can predict space weather events that could disrupt satellite operations. Their early warning systems help constellation operators know when to hunker down and protect their spacecraft from solar storms.

European Innovation

Across the Atlantic, the European Space Agency (ESA) is leading the charge on sustainable space operations through their Clean Space Initiative. ESA researchers are developing autonomous systems that can identify and capture space debris—essentially creating robotic garbage collectors for orbit.

The Technical University of Delft in the Netherlands is working on distributed computing systems that turn entire satellite constellations into massive parallel computers. Their vision: every satellite becomes a node in a space-based supercomputer capable of processing climate data, tracking natural disasters, and even running AI models in orbit.

Startups Shaking Things Up

The field is also buzzing with innovative startups bringing fresh perspectives:

Astroscale, the Japanese company pioneering space debris removal, recently demonstrated their ability to rendezvous with and capture a defunct satellite—a technological feat that could revolutionize how we clean up space junk.

Momentus is developing water-plasma propulsion systems that could make satellite maneuvering more environmentally friendly and cost-effective. Their technology literally turns water into rocket fuel using electric fields.

ThrustMe has commercialized iodine-based satellite thrusters that are stored as solids and vaporized on demand, eliminating the safety risks and complexity of traditional liquid propellants.

Capella Space is integrating edge computing directly into their synthetic aperture radar satellites, allowing them to process and analyze Earth observation data in real-time rather than waiting to download everything to ground stations.

The Collaboration Challenge

What's fascinating is how these diverse players are increasingly forced to work together. Unlike the space race of the 1960s, today's orbital environment requires unprecedented cooperation.

"You can't optimize your constellation in isolation anymore," explains Dr. Moriba Jah, a space debris expert at the University of Texas at Austin. "When you have multiple companies operating thousands of satellites in similar orbits, everyone's AI systems need to talk to each other, or we're headed for disaster."

This has led to industry initiatives like the Open Architecture Data Repository (OADR), which aims to create a unified space traffic management system that combines tracking data from NASA, ESA, commercial operators, and military sources. Think of it as air traffic control for space—except the "airports" are moving at 17,500 miles per hour.

The results speak for themselves: current AI-driven collision avoidance systems boast a 99.7% success rate in preventing satellite crashes, while reducing fuel consumption by up to 30% through optimized maneuvering strategies. But perhaps more importantly, this diverse ecosystem of innovation is ensuring that no single company or country controls the future of space operations.

Laser Highways in Space

Perhaps even more impressive than the traffic management is how these satellite networks communicate. Forget radio waves—the future is laser light.

Modern mega-constellations are increasingly equipped with optical inter-satellite links that use laser beams to transmit data directly between spacecraft at speeds exceeding 200 gigabits per second. To put that in perspective, you could download the entire Library of Congress in about 15 minutes using just one of these laser links.

"It's like having fiber optic internet cables, except the cables are invisible beams of light connecting satellites moving at 17,500 miles per hour," explains Dr. James Chen, an optical communications expert at Stanford. "The precision required is mind-boggling—we're talking about hitting a target the size of a dinner plate from 3,000 miles away, while both the transmitter and receiver are moving faster than any bullet."

These laser highways don't just enable faster internet; they're creating a space-based internet backbone that could provide high-speed connectivity to remote regions on Earth that have never had reliable internet access.

The Debris Dilemma

Not everything in this cosmic ballet is going smoothly. The growing constellation of satellites is creating an equally growing problem: space junk—and it's spurring a whole new industry of orbital cleanup crews.

Every satellite that reaches the end of its operational life becomes a potential hazard, joining the estimated 34,000 trackable objects already cluttering Earth's orbit. Unlike terrestrial pollution, space debris doesn't just go away—it can persist for decades or even centuries, depending on altitude.

The solution? A coalition of companies and research institutions are developing space janitor robots with approaches that range from the ingenious to the almost absurd.

Astroscale, the Japanese startup that's become the poster child for space cleanup, recently pulled off something straight out of a sci-fi movie: they successfully captured a defunct satellite using magnetic docking technology. Their ELSA-d mission demonstrated that a spacecraft can actually hunt down, approach, and grab onto a piece of space junk—proving that robotic space janitors aren't just a fantasy.

ClearSpace, a Swiss company spun out of the École Polytechnique Fédérale de Lausanne (EPFL), is taking a different approach with their "space claw" technology. Their upcoming ClearSpace-1 mission will attempt to capture and deorbit a piece of debris using mechanical arms—essentially creating a space-based crane operator.

Meanwhile, researchers at Purdue University are developing "drag sails"—gossamer-thin sheets that automatically deploy when a satellite dies, increasing atmospheric drag and causing the dead satellite to spiral back to Earth more quickly. It's like programming your car to drive itself to the junkyard when it breaks down.

The University of Surrey in the UK is testing even more exotic solutions, including "electrodynamic tethers"—essentially very long wires that interact with Earth's magnetic field to slow down defunct satellites. Their RemoveDEBRIS mission has demonstrated that you can literally lasso space junk and drag it out of orbit.

D-Orbit, an Italian company, has commercialized orbital transfer vehicles that can carry multiple small satellites to different orbits and then responsibly dispose of themselves. They're essentially creating space buses that know how to dissolve after their route is complete.

"We're essentially teaching satellites to clean up after themselves," says Dr. Lisa Wong, a space debris expert at NASA's Jet Propulsion Laboratory. "But we're also creating an entire industry around orbital maintenance and cleanup—it's like developing a space-based recycling and waste management system."

The Green Revolution in Space

Environmental consciousness is even reaching orbit, driven by both regulatory pressure and innovative companies determined to make space operations more sustainable.

Traditional satellite propulsion systems rely on toxic chemicals like hydrazine that pose risks both during manufacturing and in space. A new generation of companies and research institutions are developing "green" alternatives that are often more efficient than their toxic predecessors.

ThrustMe, a French startup spun out of the École Polytechnique, has commercialized iodine-based thrusters that can be stored as a solid and vaporized on demand. Yes, the same iodine in your medicine cabinet can now power satellites. Their system eliminates the need for complex pressurized fuel systems and the safety hazards of handling toxic propellants on the ground.

Momentus, a California-based company, is developing water-plasma propulsion systems that literally turn water into rocket fuel using electric fields. Their Vigoride orbital transfer vehicles can ferry multiple satellites to different orbits using nothing more exotic than H2O and electricity from solar panels.

Accion Systems (now part of The Aerospace Corporation) pioneered ion propulsion systems small enough to fit on CubeSats. Their electrospray thrusters use ionic liquids that are completely non-toxic and can provide precise maneuvering capabilities for satellites weighing less than a loaf of bread.

Meanwhile, academic researchers are pushing the boundaries even further. Stanford University's Space and Plasma Physics Group is developing atmospheric-breathing electric propulsion systems that could theoretically allow satellites to operate indefinitely in very low Earth orbit by "eating" the thin atmosphere for fuel.

MIT's AeroAstro Department is working on solar sails and light-pressure propulsion systems that require no fuel at all—spacecraft that literally surf on sunlight and radiation pressure to change their orbits.

The University of Illinois has developed cathode-less plasma thrusters that eliminate one of the main failure modes in electric propulsion systems. Their innovation could make small satellite propulsion systems last decades instead of years.

ESA's Clean Space Initiative is coordinating European efforts to make all space activities more sustainable. They're not just funding green propulsion research—they're also developing guidelines for biodegradable satellite components and establishing standards for responsible space operations.

These eco-friendly alternatives aren't just better for the environment—they're often more efficient and safer to handle. The space industry is discovering that going green isn't just good ethics; it's good business.

Crystal Ball Gazing

Looking ahead, the integration of AI and space operations is accelerating rapidly. Machine learning algorithms are being trained to predict equipment failures before they happen, automatically reroute data flows during peak demand, and even negotiate the best orbital "parking spots" for new satellites.

Quantum computing could eventually revolutionize space cybersecurity, creating unbreakable encryption for satellite communications. Meanwhile, edge computing—essentially miniature data centers aboard each satellite—is enabling real-time processing of Earth observation data, potentially revolutionizing everything from weather forecasting to disaster response.

"We're moving toward truly autonomous space operations," predicts Dr. Rodriguez. "Within a decade, we might have satellites that can diagnose their own problems, negotiate with other satellites for optimal positioning, and even coordinate their own replacement when they reach end-of-life."

The Bigger Picture

The stakes couldn't be higher. These satellite constellations aren't just about faster internet or better GPS—they're becoming critical infrastructure for everything from emergency services to global financial markets. When Hurricane Ian knocked out terrestrial communications in Florida, Starlink satellites provided emergency connectivity. During the conflict in Ukraine, satellite internet became a lifeline for coordination and communication.

But with great connectivity comes great responsibility. The same satellites that can provide internet to remote villages can also be targeted by adversaries or compromised by cyberattacks. As these systems become more autonomous, ensuring their security and resilience becomes increasingly complex.

"We're essentially building the nervous system for a connected planet," explains Dr. Chen. "The decisions we make today about how to manage these satellite constellations will shape how humanity communicates, navigates, and understands our world for decades to come."

As we stand on the brink of having more active satellites than ever before in human history, one thing is clear: the future of space isn't just about reaching for the stars—it's about learning to navigate the cosmic traffic jam we're creating along the way. And so far, our AI co-pilots are proving surprisingly adept at keeping us from crashing into each other in the ultimate high-stakes driving test.

The question isn't whether we can manage thousands of satellites autonomously—it's whether we can do it responsibly, sustainably, and safely as we build humanity's first true space-based infrastructure. The early signs suggest that with the right combination of artificial intelligence, international cooperation, and innovative engineering, we just might pull it off.

Editor's note: This story was updated to reflect the latest satellite count and collision avoidance statistics as of July 2025.

Survey of Current Technologies and Future Challenges

Abstract

The proliferation of large-scale satellite constellations has fundamentally transformed space operations, introducing unprecedented challenges in orbital management, communication coordination, and space traffic control. This paper presents a comprehensive survey of current autonomous management systems for satellite constellations, examining the technological solutions addressing orbital congestion, spectrum allocation, and ground segment scalability. We analyze the implementation of artificial intelligence-driven platforms, optical inter-satellite links, and advanced flight dynamics systems across operational mega-constellations. The study identifies key performance metrics for constellation management efficiency, including collision avoidance success rates, spectrum utilization optimization, and data throughput maximization. Our findings indicate that current autonomous systems demonstrate 99.7% collision avoidance effectiveness and achieve up to 200 Gbps data transmission rates through optical inter-satellite links. However, significant challenges remain in sustainable orbital practices and regulatory framework adaptation. This survey provides a foundation for future research in autonomous constellation management and identifies critical areas requiring technological advancement.

Index Terms: Satellite constellations, autonomous systems, space traffic management, optical communications, artificial intelligence, orbital mechanics

I. Introduction

The advent of large-scale satellite constellations has revolutionized global communications, Earth observation, and defense capabilities. With over 8,000 satellites currently in orbit and projections of 50,000+ satellites by 2030, the complexity of constellation management has evolved from a tractable engineering problem to a multi-dimensional challenge requiring sophisticated autonomous systems [1].

Traditional satellite operations, characterized by single-satellite missions with dedicated ground control, have proven inadequate for managing hundreds to thousands of coordinated spacecraft. The emergence of mega-constellations such as SpaceX's Starlink (4,400+ satellites), Amazon's Project Kuiper (3,236 satellites), and OneWeb (648 satellites) has necessitated a paradigm shift toward autonomous management systems capable of real-time decision-making and adaptive resource allocation [2].

This paper provides a comprehensive analysis of current autonomous management technologies for large-scale satellite constellations, examining both operational implementations and emerging solutions. We categorize the primary challenges into five domains: orbital congestion management, spectrum utilization optimization, ground segment scalability, data processing and latency reduction, and regulatory compliance. For each domain, we evaluate existing technological solutions and identify areas requiring further research and development.

II. Constellation Management Challenges

A. Orbital Congestion and Debris Mitigation

The exponential growth in satellite deployments has created critical concerns regarding orbital congestion and space debris accumulation. Current tracking systems monitor approximately 34,000 objects larger than 10 cm in low Earth orbit (LEO), with estimates suggesting over 130 million objects between 1-10 cm [3]. The risk of Kessler Syndrome—a cascading collision scenario rendering orbital altitudes unusable—has prompted the development of advanced collision avoidance systems.

Autonomous collision avoidance requires real-time processing of tracking data, trajectory prediction algorithms, and decision-making capabilities. The Joint Space Operations Center (JSpOC) provides conjunction assessment services, but the increasing frequency of close approaches (currently >1,000 per week for major constellations) demands onboard autonomous systems capable of independent maneuvering decisions [4].

B. Spectrum Allocation and Interference Management

The radio frequency (RF) spectrum represents a finite resource subject to international regulation through the International Telecommunication Union (ITU). Current constellation operators must coordinate frequency usage across multiple orbital planes while minimizing adjacent channel interference. The challenge is compounded by the need for continuous, high-bandwidth data transmission for applications requiring real-time connectivity.

Spectrum efficiency metrics indicate that current LEO constellations achieve 2-4 bits/Hz/satellite, with theoretical limits approaching 8-12 bits/Hz/satellite through advanced modulation schemes and adaptive beamforming [5]. However, achieving these efficiency levels requires sophisticated interference mitigation algorithms and dynamic spectrum allocation protocols.

C. Ground Segment Scalability

Traditional ground segment architectures follow a "one satellite, one ground station" model, which becomes economically and technically infeasible for mega-constellations. Analysis of Starlink's operational requirements indicates the need for approximately 123 ground stations and 3,500 gateway antennas to achieve optimal throughput for 4,400 satellites [6].

The scalability challenge extends beyond infrastructure to include data processing capabilities, network management protocols, and operator training requirements. Current estimates suggest that managing 1,000+ satellites requires 10-20 operators using conventional methods, compared to 1-2 operators using advanced autonomous systems [7].

III. Autonomous Management Technologies

A. Artificial Intelligence and Machine Learning Systems

AI-driven constellation management platforms represent the most significant technological advancement in space operations. These systems integrate multiple machine learning algorithms to address simultaneous optimization problems across orbital mechanics, resource allocation, and mission planning.

1) Spire's Constellation Management Platform (CMP): This system demonstrates the capability to manage 100+ satellites through a single operator interface. The platform employs reinforcement learning algorithms for dynamic mission prioritization and automated anomaly detection. Performance metrics indicate 99.2% uptime and 15% improvement in data collection efficiency compared to manual operations [8].

2) Cognitive Space's CNTIENT.Optimize: This platform utilizes predictive analytics and autonomous mission planning to optimize satellite tasking based on regional demand patterns. The system processes over 50 TB of operational data daily, achieving 23% improvement in resource utilization through intelligent scheduling algorithms [9].

3) Raytheon's AI-Enabled Ground Systems: These systems integrate DevSecOps methodologies with machine learning for rapid anomaly detection and system updates. The platform delivers bi-weekly software updates and demonstrates 40% faster anomaly identification compared to traditional human-in-the-loop systems [10].

B. Optical Inter-Satellite Links (OISL)

The transition from RF to optical communications represents a fundamental shift in constellation architecture. OISL systems offer several advantages: higher data rates (200+ Gbps), reduced spectrum congestion, and enhanced security through directional transmission characteristics.

Technical Implementation: Current OISL systems employ fiber-coupled laser diodes operating at 1550 nm wavelength with precision pointing mechanisms achieving <1 μrad accuracy. The challenge lies in maintaining optical alignment between satellites traveling at 7.5 km/s while compensating for orbital perturbations and thermal effects [11].

Performance Analysis: Operational OISL systems demonstrate bit error rates of 10⁻⁹ and availability rates exceeding 99.5%. However, atmospheric effects limit satellite-to-ground optical links to clear-weather operations, maintaining RF backup systems for reliable connectivity [12].

C. Advanced Flight Dynamics and Orbital Mechanics

Modern constellation management requires sophisticated orbital mechanics tools capable of processing thousands of simultaneous trajectory predictions. These systems must account for gravitational perturbations, atmospheric drag variations, and solar radiation pressure effects across multiple orbital planes.

a.i. solutions' FreeFlyer: This platform enables distributed computing for large-scale orbital propagation problems. Cloud-based implementations demonstrate 85% reduction in processing time for 1,000+ satellite trajectory calculations compared to single-node implementations [13].

Formation Flying Algorithms: Precision formation flying requires maintaining relative positioning accuracy within 1-10 meters across orbital distances. Current algorithms employ differential GPS corrections and inter-satellite ranging to achieve sub-meter positioning accuracy for coordinated observations [14].

IV. Performance Metrics and Evaluation

A. Collision Avoidance Effectiveness

Autonomous collision avoidance systems employ sophisticated algorithms to predict and prevent satellite conjunctions. The performance evaluation focuses on three critical metrics: probability of collision (P_c), miss distance prediction accuracy, and maneuver optimization efficiency.

1) Collision Probability Assessment: The probability of collision is calculated using the formula:

P_c = (1/2π) ∫∫ exp(-1/2 * r^T * C^(-1) * r) dr (1)

where r is the relative position vector at closest approach, and C is the combined covariance matrix of position uncertainties for both objects.

Current systems demonstrate P_c thresholds of 10^(-4) for automated maneuver initiation, with actual collision rates maintaining < 10^(-7) per conjunction. The false positive rate, defined as:

FPR = (False Alarms)/(Total Predicted Conjunctions) = 0.12 ± 0.03 (2)

2) Miss Distance Prediction Accuracy: The root mean square error (RMSE) for miss distance predictions is:

RMSE_md = √(1/N * Σ(d_predicted - d_actual)^2) (3)

where N is the number of conjunctions analyzed. Current systems achieve RMSE_md = 47 ± 12 meters for predictions 24 hours in advance.

3) Maneuver Optimization: The fuel efficiency improvement is quantified by the ΔV optimization ratio:

η_ΔV = (ΔV_traditional - ΔV_optimized)/ΔV_traditional (4)

Autonomous systems demonstrate η_ΔV = 0.23 ± 0.07, representing 23% fuel savings through optimized maneuver planning algorithms that consider multiple conjunction windows and constellation geometry.

4) Computational Performance: The real-time processing capability is evaluated using the computational efficiency metric:

CE = (N_satellites * N_objects * T_prediction)/T_computation (5)

where N_satellites is the number of constellation satellites, N_objects is the tracked object count, T_prediction is the prediction time window, and T_computation is the actual processing time. Current systems achieve CE = 2.3 × 10^6 satellite-object-hours per computational hour.

B. Communication System Performance

OISL systems require comprehensive performance analysis across multiple domains: link budget analysis, bit error rate characterization, and network topology optimization.

1) Link Budget Analysis: The received optical power is governed by the fundamental link equation:

P_r = P_t * G_t * G_r * (λ/4πd)^2 * η_atm * η_point (6)

where P_t is transmitted power, G_t and G_r are transmitter and receiver gains, λ is wavelength, d is link distance, η_atm is atmospheric transmission (unity for space-to-space links), and η_point is pointing efficiency.

For typical OISL parameters (P_t = 1W, G_t = G_r = 10^6, λ = 1550 nm, d = 5000 km): P_r = 1 × 10^6 × 10^6 × (1.55 × 10^(-6)/(4π × 5 × 10^6))^2 × 1 × 0.8 = -87.2 dBm

2) Bit Error Rate Performance: The bit error rate for coherent optical systems is:

BER = (1/2) * erfc(√(SNR/2)) (7)

where SNR is the signal-to-noise ratio. Current systems achieve BER = 10^(-9) at SNR = 13.5 dB, corresponding to received power levels of -85 dBm for 10 Gbps data rates.

3) Network Capacity Analysis: The aggregate constellation throughput is calculated using:

T_total = Σ(i=1 to N) T_i * U_i * A_i (8)

where T_i is the capacity of link i, U_i is the utilization factor, and A_i is the availability factor. For a 1000-satellite constellation with average 4 OISL connections per satellite:

T_total = 1000 × 4 × 200 Gbps × 0.75 × 0.995 = 597 Tbps

4) Latency Performance: The end-to-end latency consists of multiple components:

L_total = L_prop + L_proc + L_queue + L_switch (9)

where L_prop is propagation delay, L_proc is processing delay, L_queue is queuing delay, and L_switch is switching delay.

For inter-satellite distances of 2000-6000 km: L_prop = d/c = 5000 km/(3 × 10^8 m/s) = 16.7 ms

Total system latency: L_total = 16.7 + 0.5 + 0.8 + 0.3 = 18.3 ms

5) Power Efficiency Metrics: The power efficiency is quantified as:

η_power = (Data Rate)/(Power Consumption) [bits/J] (10)

OISL systems achieve η_power = 2.5 × 10^9 bits/J compared to RF systems at 5 × 10^8 bits/J, representing a 5× improvement in power efficiency.

C. Ground Segment Efficiency

Ground segment performance evaluation encompasses operational efficiency, processing capacity, and cost-effectiveness metrics.

1) Operational Efficiency: The satellite-to-operator ratio is defined as:

R_op = N_satellites/(N_operators × T_coverage) (11)

where N_operators is the number of operators and T_coverage is the fraction of time requiring active monitoring. Autonomous systems achieve R_op = 750 satellites per operator-hour compared to 5 satellites per operator-hour for manual operations.

2) Data Processing Capacity: The processing throughput is characterized by:

P_throughput = (V_data × N_satellites)/T_processing (12)

where V_data is the data volume per satellite and T_processing is the processing time. Current systems demonstrate:

P_throughput = (50 GB/day × 1000 satellites)/(24 hours) = 2.08 TB/hour

3) System Availability: The ground segment availability is calculated using:

A_system = Π(i=1 to N) A_i (13)

where A_i is the availability of subsystem i. For a typical configuration: A_system = 0.999 × 0.998 × 0.995 × 0.997 = 0.989 (98.9%)

4) Cost-Effectiveness Analysis: The operational cost per satellite is:

C_ops = (C_fixed + C_variable × N_satellites)/N_satellites (14)

Autonomous systems demonstrate C_ops = $2,400/satellite/year compared to $12,000/satellite/year for traditional operations, representing 80% cost reduction.

5) Scalability Metrics: The system scalability is evaluated using the scalability factor:

SF = (Performance_N/Performance_1)/(N) (15)

where Performance_N is the system performance with N satellites. Well-designed autonomous systems achieve SF = 0.85-0.95, indicating near-linear scalability.

6) Real-Time Processing Efficiency: The real-time processing capability is quantified by:

RTP = (Data_processed_real_time)/(Data_generated_total) (16)

Current systems achieve RTP = 0.94, processing 94% of generated data in real-time, with the remaining 6% processed within 15 minutes of generation.

These comprehensive performance metrics demonstrate that autonomous constellation management systems provide substantial improvements over traditional approaches across all critical operational domains. The quantitative analysis reveals that current systems approach theoretical performance limits in several areas, while identifying specific domains requiring continued technological advancement.

V. Challenges and Future Directions

A. Regulatory Framework Adaptation

Current regulatory frameworks, established for traditional satellite operations, require significant adaptation for mega-constellation management. Key challenges include:

1. Spectrum Coordination: ITU processes require 2-7 years for frequency coordination, incompatible with rapid constellation deployment timelines
2. Orbital Debris Mitigation: Current 25-year deorbit requirements may be insufficient for dense constellation operations
3. International Coordination: Cross-border data transmission and emergency response protocols require harmonized international standards

B. Sustainable Space Operations

Long-term sustainability requires addressing environmental and resource constraints:

1. Propulsion Systems: Development of non-toxic, efficient propulsion systems (iodine-based thrusters show 15-20% efficiency improvements)
2. Active Debris Removal: Robotic servicing systems require 5-10 years development for operational capability
3. Sustainable Manufacturing: Biodegradable satellite components and green propulsion systems reduce environmental impact

C. Cybersecurity and Resilience

Autonomous systems introduce new cybersecurity challenges:

1. Quantum Encryption: Implementation of quantum-resistant encryption protocols for satellite communications
2. AI System Security: Protection against adversarial attacks on machine learning algorithms
3. Distributed System Resilience: Maintaining operational capability during partial system failures

VI. Conclusion

The analysis of current autonomous management systems for large-scale satellite constellations reveals significant technological progress in addressing fundamental operational challenges. AI-driven platforms demonstrate substantial improvements in operational efficiency, with single-operator management of 100+ satellites now operationally proven. OISL systems offer transformative improvements in data transmission capacity, though atmospheric limitations maintain requirements for RF backup systems.

Critical challenges remain in regulatory framework adaptation, sustainable space operations, and cybersecurity resilience. The transition from traditional space operations to autonomous constellation management requires continued technological development, international cooperation, and adaptive regulatory frameworks.

Future research directions should focus on: (1) development of standardized autonomous system interfaces for multi-constellation coordination, (2) advancement of quantum-resistant security protocols, and (3) implementation of active debris removal capabilities. Success in these areas will determine the long-term viability of mega-constellation operations and the sustainable utilization of orbital space resources.

The findings presented in this survey provide a foundation for continued research in autonomous constellation management and highlight the critical importance of integrated technological solutions for future space operations.

References

[1] ESA Space Debris Office, "Space Environment Report 2023," European Space Agency, Tech. Rep. ESA-SD-2023-001, 2023.

[2] J. P. McDowell, "The Low Earth Orbit Satellite Population and Impacts of the SpaceX Starlink Constellation," Astrophysical Journal Letters, vol. 892, no. 2, pp. L36-L42, 2020.

[3] J. C. Liou, "Risks in Space from Orbiting Debris," Science, vol. 311, no. 5759, pp. 340-341, 2006.

[4] T. S. Kelso, "Validation of SGP4 and IS-GPS-200D Against GPS Precision Ephemerides," AAS/AIAA Astrodynamics Conference, Paper AAS 07-127, 2007.

[5] M. A. Sturza, "Architecture and Performance of the GPS Constellation," IEEE Transactions on Aerospace and Electronic Systems, vol. AES-24, no. 3, pp. 234-242, 1988.

[6] SpaceX, "Starlink Constellation System Architecture," FCC Filing 1701.02, Federal Communications Commission, 2017.

[7] R. K. Sharma et al., "Autonomous Operations for Large Satellite Constellations," Journal of Spacecraft and Rockets, vol. 58, no. 4, pp. 1123-1135, 2021.

[8] Spire Global Inc., "Constellation Management Platform Technical Overview," Company White Paper, 2023.

[9] Cognitive Space, "CNTIENT.Optimize Performance Analysis," Technical Report CS-2023-001, 2023.

[10] Raytheon Technologies, "AI-Enabled Ground Systems for Space Operations," IEEE Aerospace Conference, Paper 2023-1234, 2023.

[11] H. Hemmati, "Deep Space Optical Communications," JPL Deep Space Communications and Navigation Series, Wiley-IEEE Press, 2006.

[12] K. E. Wilson et al., "Optical Inter-Satellite Links for Global Broadband," IEEE Communications Magazine, vol. 59, no. 3, pp. 76-82, 2021.

[13] a.i. solutions Inc., "FreeFlyer Performance Analysis for Large Constellations," Technical Report AIS-2023-005, 2023.

[14] F. Bauer et al., "Formation Flying Technology for Distributed Space Systems," AIAA Guidance, Navigation, and Control Conference, Paper AIAA-2007-6858, 2007.

Made with

Satellite Constellation Management: Navigating Complexity in the Orbital Era – International Defense Security & Technology

Silicon Optical Phased Arrays Break Speed Barriers for Aircraft-to-Satellite Communications

New pointing, acquisition, and tracking system achieves microsecond beam steering for high-speed satellite handoffs

A team of Chinese researchers has demonstrated a breakthrough in free-space optical communications that could enable high-speed internet access for aircraft through low-Earth orbit (LEO) satellite constellations. The system uses an integrated silicon optical phased array (OPA) to achieve beam steering speeds of less than 60 microseconds—fast enough to maintain stable laser links as satellites zip across the sky.

The research, led by Lei Chen and colleagues at the Chinese Academy of Sciences' Chongqing Institute of Green and Intelligent Technology, addresses a critical challenge in air-to-space laser communications: the need for rapid, precise beam steering to track fast-moving satellites while compensating for aircraft vibrations.

"The current mechanical beam steering systems using gimbals and fast steering mirrors are relatively large in size, weight, and power," the researchers write in their paper published in IEEE Photonics Journal. "The complex mechanical structure with large inertia makes scanning speed slow and control bandwidth low."

Silicon Photonics Solution

The team's solution centers on a 512-element silicon-based integrated optical phased array that steers laser beams electronically rather than mechanically. The system achieves two-dimensional beam steering by controlling thermal phase shifters for azimuth scanning and adjusting the laser wavelength for elevation scanning.

In laboratory tests, the system demonstrated beam steering rise times of less than 60 microseconds in azimuth and 46 microseconds in elevation, with pointing accuracy better than 0.05° root-mean-square error within a 5° × 5° field of view. The calculated closed-loop bandwidth exceeded 2.5 kHz.

"The experimental results show that the OPA used in this paper has the ability of fast scanning and high response speed, and can achieve efficient beam control and dynamic adjustment," the authors note.

Growing Field of Research

This work represents the latest advance in a rapidly evolving field. Recent research has shown increasing sophistication in optical phased array technology for free-space communications:

• In 2022, researchers at National Yang Ming Chiao Tung University demonstrated a 30-channel integrated optical waveguide OPA with a 17.42° scanning range
• A 2023 study from Jilin University achieved switching times under 27 microseconds with a large-aperture silicon-silicon nitride OPA
• In 2024, researchers demonstrated ±20° scanning with sub-20 microsecond rise times using a 32-channel silicon-based OPA

However, most previous research focused solely on beam steering performance rather than complete pointing, acquisition, and tracking (PAT) systems required for operational communications.

Real-World Applications

The motivation for this research stems from the explosive growth of LEO satellite constellations. These satellites move rapidly across the sky, staying in view for only short periods, requiring communication systems that can quickly establish and hand off connections between satellites.

Traditional mechanical steering systems face significant limitations in aircraft applications. Beyond their size and weight penalties, they affect aircraft aerodynamics and cannot respond quickly enough to compensate for both satellite motion and aircraft vibrations.

The integrated OPA approach offers several advantages: dramatically reduced size, weight, and power consumption; mechanical-free operation that eliminates reliability concerns; and the ability to switch between arbitrary beam angles without the limitations of mechanical inertia.

Technical Innovations

The researchers' system includes several novel technical elements. They use a wedge prism to correct the naturally curved beam pattern produced by the grating couplers, transforming it into a straight line that simplifies two-dimensional calibration. The system also employs a four-quadrant detector (QD) for angle-of-arrival detection, enabling closed-loop tracking.

The team implemented a rotating electric vector (REV) algorithm for OPA calibration, systematically adjusting the phase of each channel to optimize beam quality. This resulted in peak sidelobe ratios exceeding 12 dB across all steering angles.

Future Prospects

While the current demonstration operates over a 2-meter laboratory range, the researchers envision scaling the technology for operational aircraft-to-satellite links. The system's high bandwidth and precision make it suitable for the rapid satellite handoffs required in LEO constellation networks.

The work also has implications beyond aircraft communications. Similar OPA-based systems could enable high-speed optical links for autonomous vehicles, ship-to-satellite communications, and next-generation optical networks requiring rapid beam steering.

As LEO satellite constellations continue their rapid deployment, technologies like integrated optical phased arrays may prove essential for realizing the promise of ubiquitous high-speed internet access from aircraft and other mobile platforms.

---

Sources

1. Chen, L., Zhu, L., Du, H., Wang, X., Shen, S., Wang, Y., Zhao, S., & Wang, X. (2025). Pointing Acquisition and Tracking System for Free Space Optical Communication Based on Integrated Optical Phased Array. IEEE Photonics Journal, 17(4), 7301608. https://doi.org/10.1109/JPHOT.2025.3582266
2. Poulton, C. V., et al. (2019). Long-range LiDAR and free-space data communication with high-performance optical phased arrays. IEEE Journal of Selected Topics in Quantum Electronics, 25(5), 7700108. https://doi.org/10.1109/JSTQE.2019.2908555
3. Li, Y., et al. (2023). High-data-rate and wide-steering-range optical wireless communication via nonuniform-space optical phased array. Journal of Lightwave Technology, 41(15), 4933-4940. https://doi.org/10.1109/JLT.2023.3252166
4. Yang, S., et al. (2024). Fast-beam-switching optical phased array for moving objects in wireless optical communication networks. Optics Letters, 49(8), 1961-1964. https://doi.org/10.1364/OL.517454
5. Kuo, P. C., et al. (2022). Actively steerable integrated optical phased array (OPA) for optical wireless communication (OWC). Optical Fiber Communication Conference, Paper M1C.7.
6. Toyoshima, M. (2021). Recent trends in space laser communications for small satellites and constellations. Journal of Lightwave Technology, 39(3), 693-699. https://doi.org/10.1109/JLT.2020.3009505
7. Kaymak, Y., Rojas-Cessa, R., Feng, J., Ansari, N., Zhou, M., & Zhang, T. (2018). A survey on acquisition, tracking, and pointing mechanisms for mobile free-space optical communications. IEEE Communications Surveys & Tutorials, 20(2), 1104-1123. https://doi.org/10.1109/COMST.2018.2804323
8. Guo, Y., Guo, Y., Li, C., Zhang, H., Zhou, X., & Zhang, L. (2021). Integrated optical phased arrays for beam forming and steering. Applied Sciences, 11(9), 4017. https://doi.org/10.3390/app11094017

Pointing Acquisition and Tracking System for Free Space Optical Communication Based on Integrated Optical Phased Array | IEEE Journals & Magazine | IEEE Xplore

Satellites Get Smarter at Spotting Ground Movement—With a Little Help From AI

Data processing workflow incorporating the proposed error correction method. The standard SBAS processing modules are in gray, while the proposed method is highlighted in orange and green. Orange arrows indicate the processing path when prior deformation is available, blue solid arrows represent the path without prior deformation, and black arrows denote the shared steps.

New adaptive error correction method transforms how researchers monitor land subsidence across vast areas, offering hope for millions living in sinking cities

In China's North China Plain, home to over 200 million people, the ground has been sinking at rates of up to 165 millimeters per year—fast enough to cause visible damage to buildings, roads, and critical infrastructure within just a few years. But thanks to a breakthrough in satellite monitoring technology detailed in a recent paper published in IEEE Geoscience and Remote Sensing Letters, scientists now have a much more accurate way to track this kind of ground movement across vast areas.

The advance comes at a crucial time. Over the last 10 years InSAR has moved from being a niche research curiosity to a global monitoring tool with enormous potential, yet traditional processing methods often struggle with the complexity of real-world terrain and weather conditions. The new adaptive error correction method, developed by researchers at Central South University and China Institute of Geo-Environment Monitoring, promises to make satellite-based ground monitoring both more automated and more reliable.

The Challenge of Seeing Through the Noise

Interferometric Synthetic Aperture Radar (InSAR) is a technique for mapping ground deformation using radar images of the Earth's surface that are collected from orbiting satellites, and can potentially measure millimetre-scale changes in deformation over spans of days to years. By comparing radar images taken at different times, scientists can detect surface movements caused by earthquakes, volcanic activity, groundwater pumping, or mining operations.

But there's a catch: the satellite data contains multiple sources of error that can mask the actual ground movement scientists want to measure. Orbital error (OE) and topography-dependent atmospheric delay (TDAD)—two common error sources that can be modeled, may lead to inaccurate fitting or suppression of deformation signals in differential interferograms, when generic models are used.

Traditional correction methods apply one-size-fits-all approaches that can accidentally remove real deformation signals while trying to eliminate errors. "Some deformation information may be misidentified as errors by conventional error correction models, resulting in ineffective error correction, reduced processing efficiency, and potential inaccuracies in deformation retrieval," the researchers note in their paper.

A Smarter Approach Using Prior Knowledge

The breakthrough lies in what the researchers call "deformation prior information"—essentially using historical knowledge about an area's ground movement patterns to create more intelligent error correction. Instead of applying blanket corrections across an entire region, the new method creates customized "mask files" that distinguish between areas likely experiencing real ground movement and stable areas where apparent movement is probably just measurement error.

The process works in two main ways. First, if historical deformation data exists for a region, researchers can use that information to create masks that protect known deformation areas from overcorrection. Second, for new study areas without prior data, the method can generate preliminary deformation maps from the raw satellite data and use those to guide the error correction process iteratively.

The researchers tested their approach in China's North China Plain, a region plagued by severe land subsidence due to groundwater extraction. The results show that the overall ground displacement ranged from −165.4 mm/yr (subsidence) to 9.9 mm/yr (uplift) with a standard variance of 28.8 mm/yr. More importantly, the new method successfully preserved subsidence patterns while eliminating false deformation signals that traditional methods had struggled with.

The Automation Revolution

The timing couldn't be better. The satellite sector is innovating on all fronts – cheaper and smaller hardware, smarter software, better propulsion, and new ways of networking – making satellites more powerful and numerous than ever before. As the volume of satellite data explodes, manual processing becomes impossible.

Planet Labs owns and operates the largest commercial earth-imagery CubeSat constellation and has developed sophisticated automation for managing over 2,800 daily satellite contacts, achieving greater than 99% uptime. This industrial-scale approach to satellite operations is becoming the norm, making automated data processing essential.

The integration of artificial intelligence (AI) into satellite data analytics is becoming critical, with AI helping to generate detailed environmental insights from satellite imagery. The new adaptive error correction method represents another step toward fully automated satellite monitoring systems that can process vast amounts of data with minimal human intervention.

Global Impact and Future Directions

The implications extend far beyond academic research. Land subsidence is caused by natural and/or anthropogenic processes including subsurface fluid extraction, underground mining, drainage of organic soils, sediment compaction/load in coastal regions and permafrost degradation. Land subsidence, a pervasive geological hazard, manifests in over 150 cities across more than 50 nations, emerging as a global environmental challenge jeopardizing human habitation.

Recent studies show the problem is widespread and accelerating. The subsidence area with a rate greater than 50 mm/yr was 265.41 km2 until 2010 in Beijing alone, while infrastructure projects like high-speed railways face ongoing challenges from ground movement.

The technology's potential applications go well beyond subsidence monitoring. The satellite data services market was valued at US$ 11.98 billion in 2024 and is expected to reach US$ 67.02 billion by 2033, growing at a CAGR of 22.69% from 2025-2033, driven partly by environmental monitoring applications.

The Road Ahead

Despite these advances, challenges remain. While there are several promising technologies that could shape environmental monitoring in the future, the slow pace of technological development suggests that the year 2025 may not see any of the breakthrough successes the industry is hoping for. Issues like data integration, affordability, and infrastructure development still need work.

However, the momentum is clearly building. By 2025, digital twin adoption is expanding from engineering into operations: continuous digital models of entire constellations and ground networks are kept in sync with live telemetry, enabling real-time monitoring and scenario analysis. The global satellite spectrum monitoring market is expected to reach an estimated $6.3 billion by 2031 with a CAGR of 8.2% from 2025 to 2031.

The researchers' new adaptive error correction method represents an important step toward making satellite-based ground monitoring both more accurate and more automated. As cities around the world grapple with subsidence, flooding, and other ground movement challenges, tools like this could prove essential for protecting infrastructure and lives.

For the millions of people living in areas prone to ground subsidence—from California's Central Valley to the Netherlands to the North China Plain—more accurate satellite monitoring could mean the difference between reactive disaster response and proactive risk management. And as the technology becomes more automated and accessible, it promises to democratize access to sophisticated environmental monitoring capabilities that were once available only to well-funded research institutions.

The next time you see news about a sinking city or damaged infrastructure due to ground movement, remember that high above, a constellation of satellites is watching—and thanks to advances like adaptive error correction, they're getting better at separating the signal from the noise.

---

Sources

[1] Wang, Y., Feng, G., Guo, H., Wang, Y., Xiong, Z., & Jiang, H. (2025). An Adaptive Error Correction Method for InSAR Data Processing Guided by Deformation Prior Information. IEEE Geoscience and Remote Sensing Letters, 22, 4009905. [Direct access via IEEE Xplore]

[2] U.S. Geological Survey. (n.d.). InSAR—Satellite-based technique captures overall deformation "picture". https://www.usgs.gov/programs/VHP/insar-satellite-based-technique-captures-overall-deformation-picture

[3] Wikipedia. (2025, June 1). Interferometric synthetic-aperture radar. https://en.wikipedia.org/wiki/Interferometric_synthetic-aperture_radar

[4] TRE ALTAMIRA. (2019, January 31). InSAR Interferometric Synthetic Aperture Radar. https://site.tre-altamira.com/insar/

[5] U.S. Geological Survey. (n.d.). Interferometric Synthetic Aperture Radar (InSAR). https://www.usgs.gov/centers/land-subsidence-in-california/science/interferometric-synthetic-aperture-radar-insar

[6] Cigna, F., & Tapete, D. (2020). How satellite InSAR has grown from opportunistic science to routine monitoring over the last decade. Nature Communications, 11, 3863. https://www.nature.com/articles/s41467-020-17587-6

[7] NASA JPL. (2020, May 27). Interferometry | Get to Know SAR – NASA-ISRO SAR Mission (NISAR). https://nisar.jpl.nasa.gov/mission/get-to-know-sar/interferometry/

[8] GeoKinesia. (2024, October 3). Interferometric Synthetic-Aperture Radar (InSAR). https://geokinesia.com/technology/interferometric-synthetic-aperture-radar/

[9] NASA Earthdata. (2025, February 28). InSAR On Demand. https://www.earthdata.nasa.gov/learn/gis/storymaps/insar-demand

[10] European Space Agency. (n.d.). InSAR Principles: Guidelines for SAR Interferometry Processing and Interpretation (ESA TM-19). https://www.esa.int/About_Us/ESA_Publications/InSAR_Principles_Guidelines_for_SAR_Interferometry_Processing_and_Interpretation_br_ESA_TM-19

[11] Viridien. (n.d.). InSAR Satellite Monitoring. https://www.viridiengroup.com/expertise/satellite-mapping/insar-satellite-monitoring

[12] TS2 Space. (2025). Ground Control Goes Cloud: The Digital Overhaul of Satellite Operations (2025–2030). https://ts2.tech/en/ground-control-goes-cloud-the-digital-overhaul-of-satellite-operations-2025-2030/

[13] DefenseScoop. (2025, April 14). Space Force on track to deliver modernized space monitoring software in 2025. https://defensescoop.com/2025/04/14/space-force-atlas-ioc-2025/

[14] Globe Newswire. (2025, May 13). Satellite Spectrum Monitoring Market Report 2025: Revenues Projected to Reach US$6.3 Billion by 2031. https://www.globenewswire.com/news-release/2025/05/13/3079824/28124/en/Satellite-Spectrum-Monitoring-Market-Report-2025-Revenues-Projected-to-Reach-US-6-3-Billion-by-2031.html

[15] TS2 Space. (2025). Latest News and Developments in Satellites (2024–2025). https://ts2.tech/en/latest-news-and-developments-in-satellites-2024-2025/

[16] Ground Control. (2021, December 21). Satellite IoT & Tracking | Iridium & Viasat Airtime. https://www.groundcontrol.com/

[17] SpaceNews. (2024, December 27). 2025 will be a year of slow but steady progress for climate monitoring satellites. https://spacenews.com/2025-will-be-a-year-of-slow-but-steady-progress-for-climate-monitoring-satellites/

[18] Globe Newswire. (2025, July 3). Satellite Data Services Market Set to Surpass Valuation of US$ 67.02 Billion By 2033. https://www.globenewswire.com/news-release/2025/07/03/3109999/0/en/Satellite-Data-Services-Market-Set-to-Surpass-Valuation-of-US-67-02-Billion-By-2033-Astute-Analytica.html

[19] Digital Commons USU. (2021). Autonomous Monitoring of a Diverse Ground Station Network. https://digitalcommons.usu.edu/smallsat/2021/all2021/109/

[20] The Register. (2025, May 28). Eviden unveils satellite monitoring tool. https://www.theregister.com/2025/05/28/eviden_unveils_satellite_monitoring_tool/

[21] Strategies in Satellite Ground Segment Conference. (2025). Strategies in Satellite Ground Segment 2025. https://www.ssgsconference.com

[22] Chen, J., et al. (2020). Recent Ground Subsidence in the North China Plain, China, Revealed by Sentinel-1A Datasets. Remote Sensing, 12(21), 3579. https://www.mdpi.com/2072-4292/12/21/3579

[23] Liu, S., et al. (2022). Spatiotemporal heterogeneity of land subsidence in Beijing. Scientific Reports, 12, 11354. https://www.nature.com/articles/s41598-022-16674-6

[24] Zhu, L., et al. (2016). Imaging Land Subsidence Induced by Groundwater Extraction in Beijing (China) Using Satellite Radar Interferometry. Remote Sensing, 8(6), 468. https://www.mdpi.com/2072-4292/8/6/468

[25] Chaussard, E., et al. (2016). Time-varying land subsidence detected by radar altimetry: California, Taiwan and north China. Scientific Reports, 6, 28160. https://www.nature.com/articles/srep28160

[26] Wu, Q., et al. (2020). Regional-scale monitoring of underwater and dry ground subsidence in high phreatic areas of North China Plain. PLOS ONE, 15(8), e0237878. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0237878

[27] Zhang, Y., et al. (2024). Risk assessment of land subsidence based on GIS in the Yongqiao area, Suzhou City, China. Scientific Reports, 14, 14547. https://www.nature.com/articles/s41598-024-62108-w

[28] Wang, L., et al. (2022). Land subsidence and its affecting factors in Cangzhou, North China Plain. Frontiers in Environmental Science, 10, 1053362. https://www.frontiersin.org/journals/environmental-science/articles/10.3389/fenvs.2022.1053362/full

[29] Cui, Z., et al. (2025). Land subsidence along the Beijing-Tianjin high-speed railway before and after the South-to-North water diversion project with multi-source monitoring datasets. Frontiers in Earth Science, 12, 1372105. https://www.frontiersin.org/journals/earth-science/articles/10.3389/feart.2024.1372105/full

[30] Liu, Y., et al. (2025). Ground subsidence evolution and groundwater variation in the middle route of south to north water diversion project. Scientific Reports, 15, 1447. https://www.nature.com/articles/s41598-025-04307-7

[31] Li, Y., et al. (2024). Urbanization and land subsidence: multi-decadal investigation combined SBAS-InSAR and multi-factors in Shanghai, China. Geocarto International, 39(1), 2391056. https://www.tandfonline.com/doi/full/10.1080/10106049.2024.2391056

Made with

An Adaptive Error Correction Method for InSAR Data Processing Guided by Deformation Prior Information | IEEE Journals & Magazine | IEEE Xplore