Breaking News
Belgian IDDEA Launches Onboard AI System for Combat Vehicles to Identify Military Equipment.
Belgian defense technology company IDDEA will unveil an embedded version of its MEGA artificial intelligence identification technology at Eurosatory 2026 in Paris, bringing offline recognition of specific military equipment models directly onboard combat vehicles, air defense systems, surveillance stations, drones, and future smart munitions. The capability moves target identification and technical exploitation from connected digital devices into closed-network and fully offline environments, giving crews and operators faster access to threat data when communications are degraded, jammed, or deliberately disconnected.
The company will present the embedded MEGA system at Hall 4 - E78 during Eurosatory 2026, where its new architecture will demonstrate how onboard artificial intelligence can support the recognition, analysis, and exploitation of observed military equipment at the tactical edge. Modern forces increasingly need local AI identification functions able to support battlefield decision-making without depending on cloud services, remote databases, or vulnerable external networks.
Related News: Belgium’s IDDEA Unveils Offline AI System for Real-Time Battlefield Military Equipment Identification
Illustrative example of embedded MEGA integration on a vehicle with sensors, operator interface, and real-time identification (Picture source: IDDEA)
MEGA currently operates from a still image or a video stream. The system receives visual input, runs artificial intelligence algorithms locally, and queries a military database to generate an exploitable result. This process goes beyond basic object detection or broad classification by identifying the observed equipment at the model level and linking it to structured technical and capability-related information.
The value of this approach lies in its ability to transform raw imagery into actionable understanding. For an analyst, vehicle crew, air defense operator, drone controller, or intelligence cell, the central requirement is not only to know that an object has been detected, but to understand what it is, what weapons it may carry, how it is protected, how it moves, and what tactical role it may play on the battlefield.
This capability addresses one of the main constraints of modern defense operations: the need to identify and analyze threats without relying on external connectivity. MEGA’s database can be preloaded before a mission and used locally throughout deployment in a closed network. Once the mission is complete and the system is reconnected in a secure environment, database updates, new equipment entries, data enrichment, and additional integrations can be carried out back at base.
That offline architecture gives MEGA direct relevance for intelligence preparation, operational planning, training, surveillance, and battlefield exploitation. It allows users to retain access to a local military equipment identification capability even in disconnected, jammed, or degraded electromagnetic environments, where cloud-based services or remote intelligence support may be unavailable.
The embedded version shown at Eurosatory 2026 takes this logic further by moving AI identification closer to the sensor and the combat system itself. Once installed in an armored vehicle, reconnaissance vehicle, air defense fire unit, fixed observation station, unmanned aerial vehicle, or other sensor-equipped system, MEGA can process optical feeds locally, run its algorithms onboard, and return an identification result linked to technical and operational data in near real time.
Illustrative MEGA workflow and platform integration overview (Picture source: IDDEA)
This transition is important because AI identification is becoming a core combat function rather than a simple software enhancement. Onboard recognition can help crews and operators shorten the interval between observation, classification, threat assessment, and response. In high-intensity warfare, where drones, decoys, armored vehicles, artillery systems, and anti-tank teams may appear in dense and fast-moving target sets, the ability to identify a specific equipment model quickly can shape both survivability and lethality.
For combat vehicles, embedded AI identification could support crews by recognizing enemy tanks, infantry fighting vehicles, anti-tank missile teams, loitering munitions, or unmanned aerial vehicles through electro-optical and thermal sensor feeds. A vehicle that can locally identify a threat and immediately associate it with known weapon range, armor layout, mobility, or tactical role gives its crew more time to maneuver, conceal, jam, or engage. This makes onboard AI recognition directly relevant to vehicle survivability, especially in drone-saturated battlefields.
For air defense systems, the same logic applies to the classification of aerial threats. AI-assisted recognition can help operators distinguish between drones, aircraft, helicopters, cruise missiles, and possible decoys, while also supporting prioritization when multiple targets are detected at the same time. In saturation attacks, accurate identification can help preserve interceptors, reduce engagement errors, and improve the efficiency of short-range and medium-range air defense networks.
The technology also has future relevance for smart ammunition and loitering munitions, where onboard target recognition could contribute to terminal identification, mission validation, and target matching. In GPS-denied or communications-contested environments, an AI-enabled munition able to compare observed imagery with stored equipment signatures could improve precision and reduce dependence on continuous operator guidance. This does not remove the need for human command authority, but it strengthens the ability of weapon systems to operate with greater resilience at the edge.
MEGA’s main operational contribution is therefore not limited to recognition alone. Its strength lies in connecting identification with contextual data. Detection tells the user that an object is present; classification may indicate that it is a tank, artillery system, drone, or air defense vehicle. Model-level identification adds the decisive layer by explaining exactly what has been observed and why it matters tactically.
This distinction is critical for intelligence and combat operations. Recognizing a main battle tank is useful, but identifying whether it is a T-72, T-90M, Leopard 2, or M1 Abrams can affect threat assessment, targeting decisions, ammunition choice, route planning, and reporting. The same applies to air defense launchers, self-propelled howitzers, multiple rocket launchers, infantry fighting vehicles, and unmanned aerial vehicles, where model-specific characteristics can influence battlefield behavior and operational risk.
The embedded MEGA architecture also reflects a broader shift in military technology toward distributed AI at the tactical edge. Instead of sending all imagery to a remote center for analysis, armed forces are increasingly seeking systems that can process data locally, reduce bandwidth requirements, and continue operating when communications are denied. This approach supports resilience, speed, and operational continuity in environments shaped by electronic warfare, cyber threats, and contested satellite communications.
IDDEA’s embedded version also opens the way to multi-domain use. The company intends to extend the MEGA logic beyond land applications, with future developments planned for naval, aerial, drone, armament, and munitions environments. This suggests a gradual evolution from a digital identification and analysis tool toward an onboard AI capability that can be adapted to different sensors, mission sets, levels of autonomy, and operational constraints.
The strategic importance of this development lies in the compression of the sensor-to-decision cycle. A combat vehicle that can identify an incoming drone or anti-tank threat gains seconds that can determine whether it survives. An air defense system that can distinguish a real threat from a decoy can use its missiles more efficiently. A smart munition able to recognize a target more accurately during the final phase of flight can improve precision in complex terrain.
As modern warfare generates more imagery, more targets, and more ambiguity, the ability to identify specific military equipment models locally will become increasingly valuable. IDDEA’s embedded MEGA technology points to a future in which AI identification is integrated directly into combat vehicles, air defense systems, unmanned systems, surveillance equipment, and precision weapons, making recognition and technical understanding part of the operational loop itself.
Visitors wishing to discover the solution and meet the IDDEA team will be able to find them at Eurosatory 2026 in Hall 4 - E78. For more information on MEGA technology or to arrange a discussion during the exhibition, contact