The Deployment of Autonomous UAVs and UCAVs on Aircraft Carriers: Opportunities and Operational Challenges

Aircraft carrier-based operations involve some of the highest-risk flight scenarios in the history of military aviation, requiring precise control. The processes of launch, glide, recovery and deck management—traditionally carried out by manned platforms—are undergoing a fundamental paradigm shift with the introduction of electromagnetic launch systems, autonomous navigation algorithms and AI-supported logistical integration. Unmanned maritime aviation not only enables long-duration mission profiles that transcend the limits of human physiology but also dramatically increases the operational range and survivability of carrier air wings (CVW).

Historical Evolution and Programmatic Transformation: From UCAS-D to the CBARS Project

The technological foundations of modern unmanned naval aviation were laid under the United States Navy’s UCAS-D (Unmanned Combat Air System Demonstration) programme. In July 2007, Northrop Grumman’s X-47B Pegasus platform secured the UCAS-D development contract, beating Boeing’s X-45C design. The first prototype of the X-47B (AV-1) made its maiden flight in February 2011 and was subsequently subjected to extensive aircraft carrier integration trials at Edwards Air Force Base and the Patuxent River Naval Air Station. Featuring a tailless, cranked flying wing geometry, the X-47B offers high aerodynamic stability at subsonic speeds, whilst its foldable wing mechanism enables high space efficiency on the flight deck, with a spot factor of 0.87 in confined parking areas. The platform made history as the first unmanned system to perform autonomous take-off from an aircraft carrier and a cable-assisted landing, powered by its Pratt & Whitney F100-PW-220U engine.

As seen in the cover photograph taken by Liz Wolter of the US Navy, the X-47B (UCAS-D) is being prepared for launch alongside the F/A-18 Hornet to conduct joint operations from the USS Theodore Roosevelt (CVN 71) (September 2014).

Table 1: Structural and Operational Specifications of the X-47B (UCAS-D) and MQ-25 (CBARS) Platforms

Platform Data / Feature X-47B Pegasus (UCAS-D) MQ-25 Stingray (CBARS)

Developer / Prime Contractor Northrop Grumman Boeing

Propulsion System (Engine) Pratt & Whitney F100-PW-220U Rolls-Royce AE 3007N (AE3700N)

Wingspan (Extended) 62.1 feet (18.9 m) 75.0 feet (22.8 m)

Wingspan (Folded) Foldable Fuselage 31.3 feet (9.5 m)

Fuselage Length ~38 feet 51.0 feet (15.5 m)

Primary Design Purpose Autonomous Combat Attack Demonstration In-flight Refuelling (Tanker)

Secondary Mission Profile: Covert and Deep Strike Reconnaissance Missions; ISR (Intelligence, Surveillance, Reconnaissance)

Targeted Range Effect: Doubling the CVW Combat Range Beyond Carrier Task Force Protection Limits

The MQ-25 Stingray (CBARS) and F/A-18 Super Hornet were spotted undergoing carrier trials on the USS George HW Bush in December 2021. Photo: Boeing

Due to fluctuations in military budgets, the cost implications of combat unmanned aerial vehicles (UCAVs) involving classified information, and shifting operational priorities, the UCLASS (Unmanned Carrier-Launched Surveillance and Strike) programme has been cancelled. The US Navy has shifted the technological gains from this programme to the CBARS (Carrier-Based Aerial-Refueling System) programme, thereby launching the MQ-25 Stingray tanker UAV project. Developed by Boeing, the MQ-25 Stingray holds the distinction of being the first mass-produced UAV to operate from an aircraft carrier. Following the signing of the engineering and production development contract in 2018, the platform’s T1 test asset prototype made its maiden flight in 2019 and successfully conducted in-flight refuelling of F/A-18F Super Hornet, E-2D Advanced Hawkeye and F-35C Lightning II aircraft in 2021.

The industrial competition between General Atomics and Boeing within the MQ-25 programme has also shaped the selection of critical component suppliers for aircraft carrier operations. General Atomics’ proposal included the Pratt & Whitney PW815 high-bypass commercial engine, UTC Aerospace landing gear, SATCOM from L3 Technologies and cyber security software from BAE Systems; the winning Boeing design, meanwhile, incorporates the Rolls-Royce AE 3007N engine and a Cobham-manufactured in-flight refuelling pod (ARS). Production of the MQ-25 is taking place at Boeing’s new 300,000-square-metre facility in Mascoutah, Illinois. Having completed its first operational production model flight on 25 April 2026, the Stingray reached Milestone C on 19 May 2026, thereby securing Low-Rate Initial Production (LRIP) approval. The MQ -25 platform is also being developed in a land-based variant (LBV) with a wingspan of 92 feet, designed to provide fuel support to CCAs (Collaborative Combat Aircraft).

EMALS and AAG Integration: Physical, Electromagnetic and Structural Limitations

EMALS (Electromagnetic Aircraft Launch System), which replaces traditional steam catapults on aircraft carriers, and AAG (Advanced Arresting Gear), which replaces hydraulic arresting systems, are setting new standards at both the physical and software levels for the integration of unmanned aircraft. Unlike conventional steam-powered systems, EMALS can dynamically adjust the launch acceleration within milliseconds according to the aircraft’s weight and the current wind speed. This prevents the structural limits of lightweight UAVs from being exceeded and reduces structural fatigue, thereby extending their service life. Similarly, the AAG system, which operates using electric motors and water turbines, continuously monitors the hook tension of the autonomously landing aircraft to apply the optimum deceleration profile.

The physical compatibility of unmanned systems with launch and arresting systems requires the development of advanced mechanical subsystems. For example, the launch bar on the nose landing gear, which is attached to the catapult carriage during launch, and the tailhook, which holds the cables during landing, are equipped with hydraulically or electrically operated actuators that can be controlled autonomously. The systems’ manoeuvrability and stability under wind loads on the deck are continuously verified by ground control stations prior to launch and landing.

One of the greatest challenges on an aircraft carrier is the intense electromagnetic interference (EMI) environment generated by EMALS and AAG. The high-voltage currents and switching circuits generated during electromagnetic launches cause powerful radio-frequency interference on the aircraft carrier. To ensure that the unmanned aircraft’s sensitive navigation receivers, autonomous control computers and data links are not affected by this environment, extensive testing is carried out in the anechoic chamber (anechoic chamber) and electromagnetic compatibility (EMC) tests are carried out at the NAS Patuxent River facility. The MQ-25’s avionics architecture and airframe design have been reinforced with specialised Faraday cage shielding and cybersecurity layers to withstand these high levels of EMI.

Autonomous Take-off and Landing Algorithms: Multi-Sensor Fusion and Ship Motion Estimation

For an unmanned aircraft to land autonomously on a moving aircraft carrier must be able to perform relative position estimation to the centimetre level. Among the leading traditional navigation solutions are Differential GPS (DGPS) and the military JPALS (Joint Precision Approach and Landing System) architecture—which are among the leading traditional navigation solutions—generate real-time relative navigation (real-time kinetic) data by processing the phase differences between the UAV and the high-precision reference stations located on the aircraft carrier. During flight tests conducted in the air, it was confirmed that DGPS navigation solutions operated with a mean root mean square error (MRSE) of 9.5 cm in total across the horizontal and vertical axes.

However, in high-intensity combat environments, the system’s redundancy is vital in situations where GPS signals are subject to enemy jamming or where the tanker aircraft’s fuselage obstructs the receiving aircraft’s line of sight to the satellite (signal blockage), the system’s redundant operation is of vital importance. In such ‘GPS-Denied’ scenarios, the high-precision Strapdown Inertial Navigation System (INS) on board the aircraft and the autonomous Stereo Machine Vision (SMV) sensors are integrated via the Extended Kalman Filter (Extended Kalman Filter – EKF). As single-camera (monocular) systems are unable to provide depth and scale information and require expensive active infrared beacons (such as VisNav beacon systems) on the receiving aircraft, SMV systems—in which two cameras are positioned at a fixed distance (baseline) behind the tanker aircraft or aircraft carrier—are preferred.

To ensure a better understanding of the subject, it would be useful to define the key terms used in the text.

Stereo Machine Vision (SMV) is an imaging technology that enables a system (in this case, an autonomous aerial vehicle) to perceive depth and distance in a manner similar to the human eye.

Single-camera (monocular) systems perceive the world as a two-dimensional ‘photograph’; consequently, they struggle to estimate how far away an object is or what its true dimensions are. Stereo Machine Vision, on the other hand, mimics the principle of biological vision to overcome this limitation.

How Does It Work?

An SMV system consists of two cameras positioned at a specific and fixed distance apart (known as the ‘baseline’).

-Parallax Effect: Just like the human eye, the two cameras view the same scene from slightly different angles.

-Depth Calculation: The system compares the ‘pixel shift’ (disparity) between the two images. The greater the difference in the position of objects within the images, the closer the object is.

-3D Modelling: Algorithms based on this difference generate a 3D map of the surroundings; thus, the aircraft can calculate its distance to the tanker aircraft without GPS, simply by mathematically processing the visual data (using the triangulation method) with no margin of error.

Why is it Critical in a Combat Environment?

The reasons for preferring SMV in autonomous systems are as follows:

1. Independence (GPS-Denied): When the enemy jammers GPS signals or communication with satellites is lost, the aircraft can continue to fly by ‘seeing’ with its own cameras.
2. Passive Operation: Infrared markers (such as VisNav) can ‘give away’ the aircraft carrier or tanker aircraft to the enemy through the light they emit. SMV systems are entirely passive; as they operate solely by utilising ambient light without transmitting any external signal, they are much more difficult to detect.
3. Accuracy: In operations requiring highly precise manoeuvres, such as autonomous refuelling, millimetre-level distance information is of vital importance. SMV provides this data in real time directly from visual data (by fusing it with the INS via the EKF).

In summary, Stereo Machine Vision is one of the most critical components of sensor fusion, enabling the aircraft not only to ‘see’ the world around it but also to ‘measure’ it.

The Extended Kalman Filter (EKF) is one of the most fundamental mathematical algorithms used in aviation, robotics and autonomous systems to perform “estimation under uncertainty”.

In the context of this text, the EKF acts as a “referee”, fusing data from the “seeing” system (camera/SMV) with the “sensing” system (Inertial Navigation System/INS), to determine the aircraft’s position and velocity with the highest possible accuracy.

Why the “Extended” Kalman Filter?

The standard Kalman Filter works brilliantly only in linear systems (i.e. for objects with constant speed and a straight path). However, aircraft:

- Their turning manoeuvres are far from linear,

- They accelerate and decelerate (they move with acceleration),

- Their angle of view is constantly changing.

It is impossible to handle these non-linear movements using the standard Kalman Filter. To overcome this challenge, the EKF uses a mathematical method known as the Taylor Series expansion, which breaks down complex movements into ‘instantaneous small linear segments’ and calculates the system as if it were linear at that moment.

How Does It Work? (Simplified Logic)

The EKF operates within a two-stage cycle at every moment:

1. Prediction: The INS (inertial navigation system) calculates where the aircraft should be in the next millisecond using its own data. However, the INS ‘drifts’ over time, meaning the error accumulates.
2. Update: The SMV (camera system) observes the external environment and measures the aircraft’s actual position (for example, its distance from the tanker aircraft).
3. Error Correction: The EKF compares the INS’s ‘prediction’ with the SMV’s ‘observation’. If there is a discrepancy between them, it minimises this error through mathematical weighting.

Why is the EKF Used in Combat?

-Data Fusion: By combining the erroneous or noisy data from multiple sensors (camera, accelerometer, gyroscope), it produces position information that is cleaner and more reliable than could be obtained from a single source.

-Noise Filtering: It filters out noise in sensor data (particularly vibration in cameras or interference in inertial sensors) .

-Speed: It has a very high computational speed, enabling it to update the position hundreds of times per second (at a high frequency).

Thanks to this integration, the aircraft filters out the ‘dirty’ data from its sensors and can achieve the millimetre-level precision required whilst the tanker aircraft is approaching.

These algorithms are combined using “sensor fusion”, a method of integrating sensor data. The relationship between sensor fusion and the Extended Kalman Filter (EKF) can be summarised in the following three main points:

-Primary Objective (Data Integrity): Sensor fusion is the process of combining erroneous, noisy or incomplete data from different sources (camera, accelerometer, GPS, etc.) using mathematical models to create a “truth data” that is more precise and reliable than could be obtained from a single sensor.

-Mathematical “Referee”: The EKF acts as a controller in this process. It assigns a ‘confidence score’ (covariance) to each sensor. For example, when image quality deteriorates, it reduces the camera’s reliability and assigns greater weight to the data from the inertial navigation system (INS).

-Error Management (Bayesian Logic): The system operates continuously in a ‘Predict and Correct’ cycle. It predicts the next position based on the previous data (Prediction), compares this with the new measurement (Update) and, if there is a difference (error), mathematically minimises this to prevent the system from drifting.

Sensor fusion defines “what is being done”, whilst the EKF defines “how this is mathematically managed flawlessly”. Together, they form a vital mechanism in aviation that stabilises the aircraft’s position in real time with millimetre-level precision.

During the final phase of landing, the six degrees of freedom (6-DOF) directly affect landing safety. To ensure that autonomous landing controllers can operate without delay, predictive algorithms are used to forecast the ship’s future movements 5 to 20 seconds in advance.

Table 2: Evaluation of Deck Motion Prediction Algorithms for Autonomous Landing and Take-off

Prediction Methodology Prediction Horizon Mathematical Input Parameter Operational Advantages / Limitations Analysis

Minor Component Analysis (MCA) Up to 20 seconds Past 6-DOF time series data Error accumulation does not increase linearly over the prediction window; it is stable; it has high computational complexity.

[Auto-Regressive (AR) / Auto-Regressive Time Series] Short-term (a few seconds) Deck acceleration and position data Computational load is extremely low, suitable for real-time operation; sensitivity decreases during sudden wave changes.

[(Long Short-Term Memory) / (LSTM)] Neural Networks Medium and long-term Wave heights and 6-DOF historical data Models non-linear, complex wave-ship interactions with high accuracy; requires intensive processing power (GPU).

[Regularised (Extreme Learning Machine) ELM] / Regularised Extreme Learning Machine: Short- to medium-term; prevents overfitting using a LAR algorithm with a selected roll series based on Lipschitz coefficients and trains quickly; requires recalibration during changes in sea state.

Wave elevation data used in ship motion prediction directly enhances the effectiveness of LSTM and MCA models. When the height and frequency of approaching waves are fed into the system via sensors and CFD models, the algorithm’s ability to predict ship roll and its stable forecast duration are significantly extended.

Aerodynamic Instabilities Along the Glide Path: Stern-Wake Turbulence and the Burble Effect

One of the most challenging aspects of aircraft carrier landing operations from an aeronautical engineering perspective is the aerodynamic turbulence known as the ‘Burble Effect’, which forms at the stern of the aircraft carrier and is a key factor in naval aviation ‘Burble Effect’. The ship’s large hull structure, flight deck and, in particular, the island superstructure positioned on the starboard side disrupt the wind flow, leading to boundary layer separation at the stern, speed losses and strong vertical winds (downwash). When a UAV on the glide path enters this flow region, it experiences a sudden drop in lift and is subjected to unplanned vertical dives. Even a very slight deviation in vertical position on the flight deck results in a lateral drift of 15:1 at the touchdown point.

An analysis of historical data reveals that, in Urnes and his team’s 1979 study, an F-4J Phantom II fighter jet passing through the aircraft carrier’s turbulence zone was recorded as having dropped vertically by approximately 1.8 metres and deviated horizontally by 38 metres from the ideal landing point. Statistics from the US Navy dating back to 1964 also confirm that 80 per cent of aircraft carrier landing accidents and 25 per cent of go-arounds (bolters/wave-offs) occurred due to insufficient vertical control.

In order to prevent these aerodynamic deviations, hybrid real-time wind prediction models utilising Computational Fluid Dynamics (CFD) simulations alongside Backpropagation (BP) Artificial Neural Networks are being implemented. The multi-dimensional wind data set, solved using the RANS and DES-SST (Detached Eddy Simulation – Shear Stress Transport) methods during the CFD phase, is utilised in the training of the BP neural networks. The instantaneous wind speed and direction (X1, X2) measured by the Doppler Wind Lidar (DWL) on board the vessel are fed into the neural network as input. Although it is challenging for the network to model the complex vortex oscillations and unstable fluctuations in the vertical wind components, the trained BP model is capable of predicting the 3D flow field with high correlation within the first 200 metres of the approach path. On the physical aerodynamic design side, passive flow control plates (stern plates) integrated into the stern region delay flow separation, thereby dampening the severity of the burble effect.

Table 3: Methodological Assessment of Wind Components in the Aircraft Carrier Approach Corridor

Wind Component BP Model Correlation Coefficient (0–200 m) Resolution (Spatial/Temporal) Operational Risk and Impact Analysis

Headwind 0.95 3 m / 3 Hz Direct effect on approach speed; sudden drops in airspeed trigger the risk of stalling.

Crosswind 0.91 3 m / 3 Hz Deviations from the runway centreline; requires horizontal rotational correction during the glide.

Vertical Wind 0.82 3 m / 3 Hz Downwash effect behind the flight deck; causes loss of lift and ‘bolter’ incidents.28

Computer Vision and Flight Deck Management: NATOPS Signal Recognition Technologies

The aircraft carrier flight deck is one of the world’s most hazardous working environments, characterised by its confined physical layout, high-wind open-sea conditions, and the intense noise and heat generated by jet engines. For unmanned aircraft to be integrated into this ecosystem, it is essential that ship’s personnel (Yellow Shirts) are able to communicate using standard hand signals. To facilitate this transition during the MQ-25 Stingray trials, a Portable Deck Control Device (HCD/DCD) was initially integrated. A Deck Handling Operator (DHO), standing immediately beside the Yellow Shirt handler, relays the hand signals received from the Yellow Shirt to the DCD unit via a screen mounted on their arm and a joystick held in their right hand. These commands autonomously control the aircraft’s nose wheel steering, wing folding mechanisms, hook release and launch bar.

Mini glossary

-[HCD: Handheld Control Device]

-[DHO: Deck Handling Operator]

To fully automate and make deck operations unmanned, DARPA and the Royal Navy are working on computer vision-based direct hand signal recognition systems. The NATOPS (Naval Air Training and Operating Procedures Standardisation) military movement database, developed in this context, contains official standards for manoeuvres on the flight deck. The database contains 24 different body and hand movements, which are very similar to one another. For example, as very small variations—such as the orientation of the palm or whether the thumb is pointing up or down—represent entirely different commands, the classification process is extremely precise.

The greatest physical constraint in the system design is the 50-foot (15.24 metres) safety zone, established to protect personnel on the flight deck from jet engine exhausts and moving aircraft. As the spatial resolution of 3D depth cameras drops dramatically at a distance of 50 feet, traditional static point cloud analysis proves inadequate. The architecture developed to overcome this obstacle utilises Motion History Images (MHI) algorithms that track dynamic motion characteristics. Images from the Bumblebee 2 stereo camera, with background noise and shadows filtered out using a codebook approach, are processed within a multi-hypothesis Bayesian inference framework. Whilst the 3D upper-body position is estimated via a skeletal generative model comprising 6 body segments and 9 joints, hand pose is determined using HOG (Histograms of Oriented Gradients) features and multi-class Support Vector Machines (SVM).

The Manned-Unmanned Teaming (MUM-T) Concept and Adversarial-Resilient Data Links

Manned-Unmanned Teaming (MUM-T), which lies at the heart of future air combat concepts, combines the tactical intelligence of manned platforms with the expendability and long-endurance capabilities of unmanned systems.11

Manned-Unmanned Teaming (MUM-T); a military operations concept that describes the synchronised, AI-supported operation of manned platforms (fighter aircraft, helicopters or command vessels) and unmanned systems (UAVs, UCAVs, UASs or autonomous ground vehicles) within the same mission ecosystem via a shared data link.

How Does MUM-T Work?

In traditional systems, UAVs are controlled from a ground control station. Under the MUM-T concept, however, tactical control of the UAV or direct management of sensor data is transferred to the pilot’s cockpit. Whilst flying the aircraft, the pilot also sends commands to the autonomous UAVs (Loyal Wingman).

What Are the Operational Advantages?

-Risk Mitigation (Force Multiplier): Unmanned platforms are sent ahead into areas where enemy air defence networks are dense and where the risk to human life is extremely high (for example, in SEAD/DEAD missions). The pilot, meanwhile, remains behind in a safe area, managing the operation.

-Sensor Fusion and Situational Awareness: UAVs transmit real-time target data, obtained via radar and electro-optical cameras, to the pilot. This allows the pilot to detect the enemy without activating their own aircraft’s radar and revealing their position.

-Cost-Effectiveness: By adding relatively cheaper and expendable autonomous UAVs alongside the very expensive fifth- and sixth-generation manned fighter aircraft, the fleet’s firepower is enhanced.

In simulator laboratory tests conducted by Boeing, it was demonstrated that an F/A-18 Super Hornet pilot could, independently of the ground station on the aircraft carrier and directly via the interface in their own cockpit, issue commands to autonomously deploy the drogue (refuelling hose basket) and initiate the refuelling process for an MQ-25 Stingray in flight.

The secure distribution of command and control in MUM-T operations is managed by the Unmanned Air Warfare Centres (UAWC) located on the aircraft carrier, which house the MD-5E Ground Control Station (GCS) software and hardware infrastructure. The backbone of this control centre, first installed on the CVN 77 (USS George H.W. Bush), the backbone of this control centre is formed by the MDCX (Multi-Domain Combat System) command and control software, developed by Lockheed Martin Skunk Works using an open system architecture. MDCX enables a single operator to simultaneously manage multiple unmanned vehicles across air, land and sea domains.

At the data link layer of this complex network architecture, the Link 16 military communications protocol serves as the primary communication tool, offering high protection against jamming and enabling secure tactical data sharing through its frequency-hopping structure. For the transmission of real-time video and radar payload data, which require high bandwidth, TCDL (Tactical Common Data Link) systems are utilised. However, operational disruptions and cyber security vulnerabilities in TCDL data links are among the factors that can limit the operational effectiveness of unmanned systems. For this reason, network security is being optimised through cyber resilience and NTCDL (Network Tactical Common Data Link) acceleration programmes.

Go-Around and Wave-Off Decision Support Mechanisms

Autonomous decision support mechanisms have been developed to address emergency scenarios that an unmanned platform may encounter whilst approaching an aircraft carrier. When the Landing Signal Officer (LSO) on the flight deck detects an obstacle on the runway or identifies that the aircraft is gliding, they press the emergency wave-off (pass-off) button. This command, which manned aircraft receive via radio announcements, is transmitted to the unmanned aircraft as a digitally encrypted signal via a direct data link. Upon receiving the signal within seconds, the autonomous flight management computer triggers the non-linear inner and outer loop flight controllers based on Dynamic Inversion (DI), switching the aircraft’s attitude and engine power parameters to those of the climb phase.

A more critical scenario is when the tail hook fails to catch any of the arresting cables on the aircraft carrier (a ‘bolter’). At this point, the ‘Bolter Logic’ software algorithm operates within the platform to prevent the aircraft from crashing into the deck and breaking up. At the moment of touchdown, when the wheels make contact with the deck, deceleration data from the accelerometers and tension sensors on the hook are scanned within milliseconds. If it is detected that the cable has not been caught and the aircraft is not decelerating, the Bolter algorithm autonomously increases engine power to the maximum military climb setting (military power / afterburner) and ensures the aircraft rapidly moves away from the deck to perform a go-around. This autonomous mechanism operates far beyond human reaction times, minimising the risk of accidents.

Digital Twin-Enabled Logistics Integration: ALIS, ODIN and Predictive Maintenance Processes

The limited capacity of maintenance workshops and spare parts storage areas on aircraft carriers make smart logistics solutions essential for the sustainability of unmanned systems. The cloud-based ODIN (Operational Data Integrated Network) architecture, which replaces the ALIS (Autonomic Logistics Information System) – which has been criticised for its high fault reporting rates and cumbersomeness – is being replaced by the cloud-based ODIN (Operational Data Integrated Network) architecture, which also forms the logistical backbone of unmanned maritime aviation. ODIN integrates with the global logistics supply chain by consolidating sensor data from the aircraft, flight logs and maintenance history into a single data lake.

The most innovative aspect of this structure is the ‘Digital Twin’ technology, designed using Model-Based Systems Engineering (MBSE) approaches. These computer-based digital replicas, fed by data from flight tests, wind tunnels and operational data, simulate the physical aircraft’s structural and hydraulic condition in real time. Boeing’s AAI (Autonomous Aircraft Inspection) and ATOM (Augmented Training Operations Maintenance) programmes, supported by 5G communications infrastructure, underpin this integration in the field. Through autonomous UAV inspections, micro-cracks and structural corrosion on the aircraft are detected with 73 per cent accuracy, enabling the creation of digital record profiles specific to each aircraft’s tail number.

During flight, the oil pressure, temperature fluctuations and vibration data from the MQ-25’s Rolls-Royce AE 3007N engine are monitored by the vehicle management computer (VMSC) and compared with the digital twin model. Should even the slightest deviation from the norm (weak signals) be detected, the system alerts ground maintenance teams before a physical failure occurs, recommending that the relevant component be replaced within a specific flight-hour window (for example, within 50 hours). This digital twin-supported predictive maintenance methodology reduces unscheduled downtime on aircraft carriers by 30 per cent, whilst also reducing unscheduled maintenance incidents by 40 per cent.

Conclusions and Future Projections

Unmanned maritime aviation has evolved from an experimental demonstration into a proven operational military doctrine, following the MQ-25 Stingray platform’s receipt of Low-Rate Initial Production (LRIP) approval and the acceleration of its integration into aircraft carriers. The structural autonomy expertise gained from the X-47B, combined with navigation fusion algorithms for autonomous take-off and landing with centimetre-level precision, enables safe flight even in the most challenging maritime conditions.

It is predicted that the proportion of unmanned systems in future aircraft carrier air wings (CVW) will exceed 60 per cent. The sustainability of this transformation depends on:

- the development of active flow control systems that model regional winds using Doppler Lidar for real-time measurement and CFD -BP networks,

-NATOPS-based autonomous deck computer vision algorithms being able to completely overcome depth constraints at ranges above 50 feet,

-Enhancing the cyber resilience of Link 16 and next-generation software-based data links,

-Ensuring that ODIN and Digital Twin-based predictive logistics processes can operate with zero delay within the global supply chain.

Consequently, unmanned maritime aviation paradigms will continue to be one of the most critical components of modern naval warfare strategies by redefining operational ranges in overseas power projection.

Our Maritime Aviation series will continue.

Reading the first seven articles that form the foundation of our Maritime Aviation series will enable you to understand the technical details and doctrinal background in this article much more clearly. I have provided the link below so that you may access the relevant publications.

Take-off and Landing Configurations on Maritime Aviation Platforms: CATOBAR, STOBAR and STOVL

https://strasam.org/savunma/deniz-silah-ve-sistemleri/deniz-havaciligi-platformlarinda-inis-kalkis-konfigurasyonlari-catobar-stobar-ve-stovl-4160

The Evolution of US Navy Jets from an Engine Architecture Perspective

https://strasam.org/savunma/deniz-silah-ve-sistemleri/motor-mimarisi-perspektifinden-abd-donanma-jetlerinin-evrimi-4166

Structural and Technical Evolution of US Navy Aircraft Carrier-Based Aircraft: 1945–1965

https://strasam.org/savunma/deniz-silah-ve-sistemleri/abd-donanmasi-ucak-gemisi-ucaklarinda-yapisal-ve-teknik-evrim-19451965-4168

Structural and Technical Evolution of US Navy Aircraft Carrier Aircraft: 1965–2025 https://strasam.org/savunma/deniz-silah-ve-sistemleri/abd-donanmasi-ucak-gemisi -structural-and-technical-evolution-of-aircraft-on-aircraft-carriers-1965-2025-4172

Confined-Space Logistics in Naval Aviation and Maintenance Engineering on Aircraft Carriers: The Process of Returning to Operational Service from the F-4 to the F-35

https://strasam.org/savunma/deniz-silah-ve-sistemleri/deniz-havaciliginda-dar-alan-lojistigi-ve-ucak-gemilerinde-bakim-muhendisligi -the-process-of-returning-to-operations-from-the-f-4-to-the-f-35-4173

Vertical Logistics of Floating Fortresses: The Elevator Systems of Nimitz and Ford-Class Aircraft Carriers

https://strasam.org/savunma/deniz-silah-ve-sistemleri/yuzen-kalelerin-dikey-lojistigi-nimitz-ve-ford-sinifi-ucak-gemilerinin-asansor-sistemleri-4178

The Evolution of Ammunition Transport Logistics on Aircraft Carriers: From the Nimitz to the New-Generation Ford-class

https://strasam.org/savunma/deniz-silah-ve-sistemleri/ucak-gemilerinde-muhimmat-tasima-lojistiginin-evrimi-nimitzden-yeni-nesil-ford-sinifina-4188

References

1. MQ-25 Stingray - Naval Drones, http://www.navaldrones.com/MQ-25-Stingray.html
2. Future Utilisation of Unmanned Aerial Vehicles Within the United States Navy - ScholarWorks, https://scholarworks.calstate.edu/downloads/m039kb782
3. How the MQ-25A Unmanned Refuelling Aircraft Could Transform the US Navy Carrier in a Future Pacific War - Army Recognition, https://www.armyrecognition.com/focus-analysis-conflicts/navy/naval-technology/how-mq-25a-unmanned-refueling-aircraft-could-transform-u-s-navy-carrier-in-future-pacific-war
4. UCLASS to be Scaled Back for CBARS Conversion, Also Known as the MQ-25 Stingray - Defense Industry Daily, https://www.defenseindustrydaily.com/uclass-to-be-descoped-for-cbars-conversion-aka-mq-25-stingray-03557/
5. GA-ASI Announces Industry-Leading Partnerships for the MQ-25 | General Atomics Aeronautical Systems Inc., https://www.ga-asi.com/ga-asi-announces-best-in-industry-partnerships-for-mq-25
6. U.S. Navy Clears Boeing MQ-25A Stingray for Production to Expand Carrier Refuelling Reach - The Defense Watch, https://thedefensewatch.com/naval-maritime/us-navy-boeing-mq25a-stingray-low-rate-production/
7. U.S. Navy Clears Boeing MQ-25A Stingray Carrier Tanker Drone for Low-Rate Production, https://www.armyrecognition.com/news/navy-news/2026/u-s-navy-clears-boeing-mq-25a-stingray-carrier-tanker-drone-for-low-rate-production
8. X -47B unmanned aircraft completes first major phase of flight-testing - New Atlas, https://newatlas.com/x-47b-flight-test-first-phase-completed/22958/
9. Unmanned Carrier Aviation - MQ-25 - NAVAIR, https://www.navair.navy.mil/product/Unmanned-Carrier-Aviation
10. Rolls-Royce powers unmanned U.S. Navy MQ-25A to successful first flight, https://www.rolls-royce.com/media/press-releases/2026/27-04-2026-rr-powers-unmanned-us-navy-mq-25-ato-successful-first-flight.aspx
11. Boeing Tests Software That Allows Super Hornet Pilot To Command ..., https://theaviationist.com/2024/05/03/new-sotware-f-18-pilot-commands-mq25/
12. 2020's - VAW VRC VUQ Foundation, https://www.vaw-vrcreadyroom.org/2020s.html
13. BAE to enhance American MQ-25A refuelling drone - UK Defence Journal, https://ukdefencejournal.org.uk/bae-to-enhance-american-mq-25a-refuelling-drone/
14. Welcome Aboard - Boeing, https://www.boeing.com/content/dam/boeing/boeingdotcom/features/innovation-quarterly/2022/05/Innovation_Quarterly_2022_Q2_042222a.pdf
15. U.S. Navy Clears Boeing MQ-25A Stingray for Low-Rate Production After Successful First Flight - The Defense Watch, https://thedefensewatch.com/aerospace-aviation/us-navy-clears-boeing-mq-25a-stingray-low-rate-production/
16. U.S. Navy Clears MQ-25 Stingray Uncrewed Tanker for Production - The Aviationist, https://theaviationist.com/2026/05/20/u-s-navy-clears-mq-25-stingray-uncrewed-tanker-for-production/
17. Boeing Unveils MQ-25's Land-Based Variant - The Aviationist, https://theaviationist.com/2024/09/16/boeing-unveils-mq-25s-land-based-variant/
18. Navy Completes Initial Carrier Demo for MQ-25 Programme, https://www.navy.mil/Press-Office/News-Stories/Article/2879506/navy-completes-initial-carrier-demo-for-mq-25-program/
19. Combining Stereo Vision and Inertial Navigation for Automated ..., https://arc.aiaa.org/doi/10.2514/1.G002648
20. Building Europe’s Satellite Navigation System, https://www.ion.org/newsletter/upload/ION-Winter2026.pdf
21. Ship-motion prediction: Algorithms and simulation results | Request PDF - ResearchGate, https://www.researchgate.net/publication/4087942_Ship-motion_prediction_Algorithms_and_simulation_results
22. Linear Velocity-free Visual Servoing Control for Unmanned Helicopter Landing on a Ship with Visibility Constraint - DR-NTU, https://dr.ntu.edu.sg/server/api/core/bitstreams/90f54d52-4acd-4aad-bcfe-288c75e7ebd2/content
23. Ship Motion Prediction Algorithms Explained | PDF | Prediction | Root Mean Square - Scribd, https://www.scribd.com/document/246260457/Ship-motion-Prediction-Algorithms-and-Simulation-Results
24. Prediction and Compensation Model of Longitudinal and Lateral Deck Motion for Automatic Landing Guidance System - MDPI, https://www.mdpi.com/2227-7390/10/19/3440
25. Search results for: SST - World Academy of Science, Engineering and Technology, https://publications.waset.org/abstracts/search?q=SST&page=4
26. CVN Airwake Modelling and Integration: Initial Steps in the Creation and Implementation of a Virtual Burble for F-18 Carrier Landing Simulations - ResearchGate, https://www.researchgate.net/publication/268556554_CVN_Airwake_Modeling_and_Integration_Initial_Steps_in_the_Creation_and_Implementation_of_a_Virtual_Burble_for_F-18_Carrier_Landing_Simulations
27. MODELLING OF AIRFLOW OVER A GENERIC AIRCRAFT CARRIER WITH AND WITHOUT AN ISLAND STRUCTURE, https://www.intmaritimeengineering.org/index.php/ijme/article/download/803/32
28. Rapid Prediction of High-Resolution 3D Ship Airwake in the Glide ..., https://www.mdpi.com/2076-3417/15/15/8336
29. Search results for: aircraft structures - World Academy of Science, Engineering and Technology, https://publications.waset.org/abstracts/search?q=aircraft%20structures&page=2
30. MQ-25 Stingray Has Begun Taxi Tests (Updated) - The War Zone, https://www.twz.com/air/mq-25-stingray-has-begun-taxi-tests
31. U.S. Navy and Boeing Complete First Carrier Tests for the MQ-25 - News Releases | Boeing Newsroom, https://boeing.mediaroom.com/news-releases-statements?item=130980
32. (PDF) Tracking body and hands for gesture recognition: NATOPS ..., https://www.researchgate.net/publication/224238166_Tracking_body_and_hands_for_gesture_recognition_NATOPS_aircraft_handling_signals_database
33. Navy completes installation of first MQ-25 Unmanned Air Warfare Centre ..., https://www.navair.navy.mil/news/Navy-completes-install-first-MQ-25-Unmanned-Air-Warfare-Center-aboard-USS-George-HW-Bush-CVN
34. U.S. Navy Installs MQ-25 Unmanned Air Warfare Centre aboard the USS George H.W. Bush Aircraft Carrier - The Aviationist, https://theaviationist.com/2024/08/18/mq-25-unmanned-air-warfare-center-uss-bush-aircraft-carrier/
35. Skunk Works®’ MDCX™ Controls Boeing MQ-25A™ First Flight - PR Newswire, https://www.prnewswire.com/news-releases/skunk-works-mdcx-controls-boeing-mq-25a-first-flight-302754358.html
36. Skunk Works®’ MDCX™ Controls Boeing MQ-25A™ First Flight - 27 April 2026, https://news.lockheedmartin.com/2026-04-27-Skunk-Works-R-MDCX-TM-Controls-Boeing-MQ-25A-TM-First-Flight
37. FY 2010 Annual Report – Director of Operational Test and Evaluation, https://www.dote.osd.mil/Portals/97/pub/reports/FY2010/other/2010DOTEAnnualReport.pdf?ver=2019-08-22-112601 -643
38. DO T &E FY 2019 Annual Report - Osd.mil, https://www.dote.osd.mil/Portals/97/pub/reports/FY2019/other/2019DOTEAnnualReport.pdf?ver=2020-01-30-115634-877
39. NDIA Defence and Industry Autumn Forum PMW 750 Overview, https://www.ndia-sd.org/wp-content/uploads/Briefs/Forum2024/Wednesday/06NDIAFallForum_PMW750_2024.pdf
40. Navy, Northrop Achieve Historic First With (Mostly) Successful X-47B ..., https://breakingdefense.com/2013/07/navy-northrop-score-historic-first-with-successful-x-47b-drone-carrier-landing/2/
41. Automated Carrier Landing of an Unmanned Combat Aerial Vehicle Using Dynamic Inversion - CORE, https://core. ac.uk/download/pdf/322583171.pdf
42. Here’s Exactly What’s In America’s Massive Advanced Weapons Deal With The UAE, https://www.twz.com/37547/heres-exactly-whats-in-americas-massive-advanced-weapons-deal-with-the-uae
43. How to Use Digital Twins for Predictive Maintenance in Aviation - ePlaneAI, https://www.eplaneai.com/blog/digital-twin-predictive-maintenance
44. The future of aerial refuelling: Accelerating automation - Eaton, https://www.eaton.com/us/en-us/markets/aerospace/knowledge-center/the-future-of-aerial -refuelling-accelerating-automation.html
45. Enhancing Aircraft Maintenance with Digitalisation, 5G and Digital Twins - Portfolio Edge, https://portfolioedge.com.au/2024/05/20/enhancing-aircraft-maintenance-with-digitalization-5g-and-digital-twins/
46. Defence News, http://www.defenceindustryreports.com/defence_news.html