The historical development of military drones, or unmanned aerial vehicles (UAVs), stretches back to the early 20th century. Initially, progress was hampered by limitations in propulsion, control, communication, and navigation technologies, coupled with a fragmented understanding of their potential applications. Development remained relatively slow until the 1990s, when the prominent use of US military drones in several high-tech local conflicts captured the intense focus of armed forces worldwide. Entering the 21st century, military drones have played an increasingly vital role in missions such as surveillance and reconnaissance, tracking of specific targets, battle damage assessment, and aerial fire support. With advancements in communication technology and artificial intelligence, and the conceptualization of manned-unmanned teaming (MUM-T), military drones are now advancing into more complex domains like air superiority and strike missions.
Unlike manned aircraft, a military drone system—particularly medium and large platforms—is a complex equipment system comprising the aerial vehicle, ground control stations (GCS), and ground support systems. It demands characteristics like agile deployment, rapid mobility, and strong battlefield survivability. Consequently, the support for military drones must be a more systematic, integrated logistics approach. It is imperative to fully leverage modern information technologies to achieve rapid diagnosis and comprehensive assessment of all drone subsystems, ultimately realizing cross-domain, precise, and agile informationalized support capabilities. This is crucial for enabling military drones to rapidly achieve full operational readiness.
Concept and Drivers of Integrated Logistics Informatization
Integrated logistics informatization refers to the application of information technology within equipment maintenance and support workflows. It involves the in-depth development of various maintenance information resources and the informatization construction or transformation of all stages, including maintenance operations, management, and supply support. This process aims to streamline support workflows, integrate information technology with equipment maintenance techniques, focus support resources on maintenance activities, enhance the precision and scientific rigor of equipment support, and ultimately improve overall equipment availability and readiness.
The development of integrated logistics informatization is primarily driven and influenced by three key factors:
- Technology Push: The continuous evolution of information technology accelerates the update cycle of information systems, consequently propelling advancements in equipment support informatization technologies.
- Support Demand Pull: Within the current maintenance framework, specific informatization capabilities are required to fulfill comprehensive support needs such as technical status management, supply support, maintenance operations, and training support for the equipment.
- Future Warfare Needs: To enhance operational effectiveness in future combat scenarios, focusing on mission/performance-based support and enabling integrated joint support, adjustments and changes in integrated logistics informatization strategies are necessary.
Current State and Challenges in Support Informatization
In the practice of military aviation integrated support, nations continuously develop their own support ecosystems. A robust and sophisticated support information system forms the foundation for the efficient operation of any military aircraft’s integrated support system.
Taking the United States as an example, through the establishment of an Integrated Logistics System (ILS), the Air Force Materiel Command plays a pivotal, albeit indirect, role in guiding equipment maintenance across units. This is evident in two main areas: first, implementing full life-cycle data management for military aircraft, including controlling technical configurations, approving technical documentation modifications, issuing technical bulletins, and adjusting maintenance intervals; second, overseeing depot-level maintenance for specified aircraft and managing the procurement, storage, and supply of parts and spares, ensuring timely support to operational units.
Since the 21st century, air forces have increasingly emphasized and promoted the transformation towards integrated logistics informatization. New platform developments have been accompanied by specialized informational support equipment, such as Interactive Electronic Technical Manuals (IETM), Portable Maintenance Aids (PMA), Integrated Training Systems (ITS), and Maintenance Support Information Systems (MSIS). These elements contribute to a relatively complete informational support architecture, significantly improving maintenance efficiency.
U.S. Military Applications and Persistent Issues
The Autonomic Logistics System (ALS) for the F-35 represents a novel maintenance and supply support concept leveraging advanced digital information technologies. Under this scheme, nearly all common testing, maintenance, and support activities for the F-35 are intended to be automated. Within the F-35’s ALS, the Prognostics and Health Management (PHM) system and the Autonomic Logistics Information System (ALIS) are its most critical functional systems and key enabling technologies. Despite significant progress in development and deployment, numerous problems persist, failing to meet all operational requirements.
During development, ALIS faced continual delays (the ALIS 3.0 version required for F-35 operational testing was released in 2018, eight years after the initial plan). It suffered from substantial system defects. High false alarm rates and data errors impaired accurate assessments of aircraft health by managers, while chaotic parts management increased the burden on maintenance personnel. The U.S. Government Accountability Office identified ALIS as a significant factor affecting the F-35’s mission readiness.
Consequently, in January 2020, the U.S. Department of Defense announced the replacement of ALIS with a new “Operational Data Integrated Network” (ODIN). Initial deployments of ODIN to F-35 squadrons were completed in July and August 2021.
Domestic Applications and Systemic Challenges
Currently, various computer software systems related to integrated support are in use within operational units. These include aviation maintenance support systems, maintenance management systems, electronic logbook systems, technical data management systems, flight data analysis software, oil analysis software, tool and equipment management software, aeronautical material management systems, and training systems.
Although numerous support information systems are fielded or under development for various aircraft platforms, they often operate in isolation or represent redundant developments. A lack of holistic planning and uncoordinated consideration of support elements leads to dispersed information. Resources remain difficult to track in real-time (“invisible”), comprehensive support situational awareness is lacking, and the level of intelligent decision support for maintenance actions is low, hampering the rapid generation of flexible and precise support plans. The following table summarizes key challenges in current support informatization.
| Challenge Area | Specific Manifestation | Impact on Support |
|---|---|---|
| System Integration | Isolated or siloed systems; redundant development. | Hinders data flow and creates information barriers. |
| Data Management | Dispersed support information; non-standardized data. | Prevents a unified view of asset health and resource status. |
| Situational Awareness | Lack of a comprehensive, real-time support common operational picture. | Impairs command and control of support operations. |
| Decision Support | Low level of intelligent analytics and predictive capabilities. | Leads to reactive, less efficient maintenance and resource allocation. |
| Resource Visibility | Inability to track personnel, spares, and equipment in real-time. | Causes delays and inefficiencies in supply chain and task allocation. |
Distinctive Characteristics of Military Drone Support
Compared to manned aircraft, military drones—especially medium and large types—exhibit unique support characteristics due to their different system composition, separation of operator and platform, and varied mission profiles.
- Separated Deployment: The Ground Control Station (GCS) and the drone platform are typically deployed separately, often in a one-to-many (one GCS controlling multiple drones) configuration. This separation exists not only between the GCS and the aircraft but also between different GCS units themselves. This necessitates coordinated support for both the GCS and the drone platforms during mission preparation phases, differing from the single-aircraft-centric support model of manned aviation.
- System Complexity: Military drones often feature long endurance, large operational radii, diverse mission payloads, and complex platform systems. This generates vast amounts of maintenance data, requiring capabilities for rapid and accurate data processing to provide timely launch decisions and fault diagnosis recommendations.
- High Mobility and Rapid Redeployment: Military drones are designed for agile deployment. Different mission types may require different configurations. The ground support system must be able to flexibly configure transportable support resources based on mission type, duration, sortie intensity, and operational conditions, aiming to minimize the logistical footprint during redeployment.
Therefore, the informatization of support for military drones must inherently enable cross-domain support (spanning geographical and system boundaries), precise support (targeted resource application), and agile support (rapidly adaptive and responsive).
Considerations on New Information Technology Development and Application
To address the support challenges of military drones and meet the demands of future large-scale, sustained operations and highly mobile, agile support, the informatization of military drone support should leverage new technologies like data links, big data, and artificial intelligence. This will enable air-ground integration, precise fault diagnosis and prediction, and intelligent support decision-making.
Integrated Support Based on Data Links
1) The Military Drone Data Link: The data link is the critical tether connecting the drone and its GCS. Its primary function is transmitting various information and commands using agreed communication protocols, enabling the GCS to precisely control the drone’s flight and other actions. Conversely, the drone transmits data to the GCS, including system status, critical consumable levels, and major fault data. Key data categories include:
- System Status Data: Airspeed, pressure altitude, throttle position, engine RPM, oil inlet temperature, etc.
- Critical Consumable Data: Fuel quantity in various tanks, battery charge state, etc.
- Severe Fault Data: Fault codes, time of occurrence, failure impact level.
Thus, leveraging the military drone data link enables real-time health status monitoring, facilitating autonomous triggering of fault warnings and responsive support tasks, shifting the paradigm from “reactive” to “proactive” or “anticipatory” support.
2) Space-Based Internet: Comprising Low Earth Orbit (LEO) satellite constellations, space-based internet offers advantages like low latency and reduced link loss, providing global or regional internet services. By accessing a secure space-based internet, drone fleets can achieve cross-domain support capabilities, enabling data sharing, resource integration, and mission coordination across vast distances.
Big Data-Enabled PHM Systems
Maintenance data from military drones exhibits the classic 5V characteristics of Big Data: Volume (large scale), Velocity (high generation speed), Variety (diverse data types from multiple subsystems), Value (low value density requiring extraction), and Veracity (data accuracy and trustworthiness). Conventional software struggles to uncover patterns and correlations within such vast, complex datasets. Therefore, utilizing cloud computing, cloud storage, and distributed database technologies, a comprehensive analysis of full life-cycle data—from design and manufacturing to in-service operations—can be performed. Through deep data mining and established fault models, the underlying physical meaning and causal relationships between features can be identified. This enables accurate and timely fault detection and isolation, remaining useful life prediction, and comprehensive health state assessment. A PHM process can be conceptually modeled as follows:
Let $S(t)$ represent the overall health state of a military drone system at time $t$. This state is a function of multiple feature vectors extracted from heterogeneous data sources:
$$ S(t) = F\Big(\vec{D}_{link}(t), \vec{D}_{post}(t), \vec{D}_{test}(t), \vec{D}_{design}, \vec{D}_{manufacturing}\Big) $$
where:
$\vec{D}_{link}(t)$: Real-time data stream from the data link.
$\vec{D}_{post}(t)$: Post-flight maintenance data download.
$\vec{D}_{test}(t)$: Ground test equipment data (e.g., radar, EW tests).
$\vec{D}_{design}$: Historical design and specification data.
$\vec{D}_{manufacturing}$: Manufacturing and assembly records.
The function $F$ represents the big data analytics and model-based reasoning process, which could involve machine learning algorithms for anomaly detection:
$$ \text{Anomaly Score } A(t) = M_{ML}(\vec{D}_{input}(t)) $$
where $M_{ML}$ is a trained machine learning model (e.g., an isolation forest, autoencoder) that compares current operational data $\vec{D}_{input}(t)$ against a learned baseline of normal behavior. A significant deviation indicates a potential fault.
Similarly, for Remaining Useful Life (RUL) prediction of a critical component, a regression model might be used:
$$ \widehat{RUL}(t) = G\Big(\vec{D}_{history}[0:t], \vec{F}_{degradation}\Big) $$
where $G$ is a predictive model (e.g., a recurrent neural network or a particle filter-based state-space model) that estimates future life based on historical data up to time $t$ and known degradation patterns $\vec{F}_{degradation}$.
| Data Category | Description & Source | Primary Application in PHM |
|---|---|---|
| Real-Time Telemetry | Continuous stream via data link (flight params, system states). | Real-time health monitoring, immediate fault detection. |
| Post-Flight Data | High-density recorded data downloaded after sortie. | Detailed trend analysis, latent fault discovery, model training. |
| Ground Test Data | Data from specialized test equipment for avionics, payloads. | In-depth system performance validation, calibration. |
| Design & Manufacturing Data | As-designed specs, tolerances, bill of materials, assembly logs. | Providing baseline for comparison, understanding failure modes. |
| Maintenance History | Records of all past repairs, part replacements, inspections. | Reliability analysis, predicting next failure, optimizing schedules. |
AI-Driven Support Decision-Making
Artificial Intelligence (AI) aims to develop machines that can mimic and execute certain human intellectual functions. Since 2011, propelled by breakthroughs in computing and data availability, AI—particularly deep neural networks—has achieved remarkable success across various fields.
By integrating AI technologies, support systems can be endowed with autonomous learning capabilities. These systems can analyze historical and current data to deduce equipment support patterns under different operational scenarios, scales, and intensities. This enables scientific prediction of material consumption, assisting in the formulation of support plans and guiding activities such as equipment allocation, supply, maintenance scheduling, and transportation. The goal is to render support decision-making more precise and data-driven. For instance, an intelligent scheduler could optimize maintenance tasks and resource allocation by solving a constraint optimization problem, minimizing total downtime or maximizing fleet availability subject to resource limits.
A simplified objective for mission-ready drone availability could be formulated as maximizing the number of operationally ready military drones:
$$ \max \sum_{i=1}^{N} R_i $$
subject to:
$$ \sum_{j \in M_i} x_{ij} \cdot t_{ij} \leq T_{avail} \quad \forall i $$
$$ \sum_{i} r_{kj} \cdot x_{ij} \leq R_{k}^{total} \quad \forall k, j $$
$$ x_{ij} \in \{0,1\} $$
where:
$R_i$: Readiness state (1=ready, 0=not ready) of military drone $i$.
$x_{ij}$: Decision variable (1 if maintenance action $j$ is performed on drone $i$, 0 otherwise).
$t_{ij}$: Time required for action $j$ on drone $i$.
$T_{avail}$: Available maintenance time window.
$r_{kj}$: Quantity of resource $k$ (e.g., a specific spare part, technician type) needed for action $j$.
$R_{k}^{total}$: Total available quantity of resource $k$.
$N$: Total number of military drones.
$M_i$: Set of required maintenance actions for military drone $i$ (predicted by PHM/AI models).
Key Capabilities for Military Drone Informatized Support
Military drone informatized support must fully utilize modern IT, establish data standards, construct big data platforms, and integrate support information systems to achieve the following key capabilities, providing powerful support for diverse military missions.
- Real-Time Support Situation Awareness: Throughout the drone’s flight, acquire system status, consumable data, and fault information via the data link. Ground-based PHM systems classify, store, and rapidly process this real-time data, performing dynamic system monitoring and health assessment. This provides maintenance crews with a visual representation of the military drone’s real-time health status and relevant flight information.
- Comprehensive System Health Evaluation: Beyond real-time data, integrate post-flight maintenance data, ground test data, and historical design/manufacturing data into a unified big data platform. Perform advanced analytics to enhance ground PHM capabilities, identifying trends in aircraft state/performance, predicting airframe and component remaining useful life, and comprehensively assessing mission system performance and operational capability.
- Intelligent Maintenance and Support Decision Support: Leverage AI techniques to enable support systems to mimic human cognitive functions—thinking, learning, reasoning—to solve or assist in solving support problems. Assess the military drone’s fitness for a given mission from both flight capability and mission capability perspectives. Automatically generate mission-tailored support plans and autonomously plan specialized mission training support.
- End-to-End Support Process Control and Optimization: Apply digital twin technology to model, simulate, and monitor the entire support process in real-time. The system should be capable of proactively identifying issues, autonomously reconfiguring and optimizing workflows, and monitoring the status and consumption of support resources with predictive alerting capabilities.
Conclusion
In the future operational landscape, military drones will play an increasingly prominent role as integral components of the combat system. To sustain the development of drone technology and enhance support effectiveness, it is essential to fully apply modern information technologies such as secure data links, big data analytics, and artificial intelligence. The development of a new generation of intelligent support information systems is paramount to realizing the vision of cross-domain, precise, and agile mission-oriented informationalized support capabilities for military drones at the earliest opportunity.