Havelsan unveils Advent-AI naval combat system to counter swarm attacks and electronic warfare.

Havelsan has unveiled Advent-AI, a new artificial intelligence layer for the Advent Combat Management System, during SAHA Expo 2026 in Istanbul, expanding the Turkish Navy’s combat architecture to help warships detect, prioritize, and respond faster to swarm attacks and electronic warfare threats. Developed with the Turkish Naval Forces’ Armerkom research center, the system strengthens decision-making inside overloaded combat information centers by filtering tactical data from radars, sonar, drones, and electronic warfare systems while keeping human operators in full control of weapon release authority.

Advent-AI introduces AI-assisted anomaly detection, predictive engagement support, and electronic warfare resilience designed to maintain target tracking and tactical prioritization even under jamming, spoofing, or degraded sensor conditions. Because the Advent CMS already equips Turkish and export naval fleets in countries including Pakistan, Indonesia, Ukraine, and Chile, the AI layer could rapidly expand through software upgrades, reinforcing a broader shift toward software-defined naval warfare, manned-unmanned teaming, and distributed maritime operations.

Related topic: Türkiye's HAVELSAN Integrates Bayraktar TB3 Drone With ADVENT CMS on TCG Anadolu Assault Ship

Havelsan's Advent-AI allows naval crews to process large volumes of combat data faster by automatically detecting abnormal threats, prioritizing targets, assisting tactical decisions, and maintaining situational awareness during electronic warfare, dense maritime traffic, and multi-threat engagements. (Picture source: Havelsan)

During the SAHA Expo 2026 in Istanbul, Turkish company Havelsan unveiled the Advent-AI, a new artificial intelligence layer for the Advent Combat Management System (CMS). Developed jointly with the Turkish Naval Forces Command’s Research Center Command, Armerkom, this AI expands the Advent architecture toward an AI-assisted naval battle management and follows Havelsan’s August 2025 integration of the MAIN artificial intelligence framework into naval systems. Demonstrated functions included anomaly detection, predictive engagement support, AI-assisted target classification, voice-assisted interaction, navigation risk monitoring, and operation under electronic warfare conditions.

Havelsan emphasized that operators retain final engagement authority and that Advent-AI remains limited to support, prioritization, and tactical assistance rather than autonomous weapon release. The Advent already equips Turkish and export naval programs, including fleets in Pakistan, Indonesia, Ukraine, and Chile, meaning the Advent-AI layer can potentially be integrated through software modernization packages. The development of the Advent CMS began during the early 2010s as a successor to the Genesis effort, with the baseline system entering operational Turkish Navy service aboard the Ada-class corvette TCG Kinaliada (F-514) in September 2019.

The Advent CMS integrates radar systems, sonar suites, electro-optical sensors, electronic warfare systems, tactical data links, missile control systems, and naval gunfire control systems into a centralized tactical-processing environment responsible for target identification, engagement sequencing, and weapon assignment. The Turkish CMS is also designed around NATO interoperability standards for coalition naval operations and multinational task groups. Unlike earlier ship-centered combat systems, the Advent was structured from the outset for distributed maritime operations involving simultaneous data exchange between multiple vessels and unmanned systems.

Advent-AI, therefore, expands an already network-oriented naval architecture rather than introducing a separate combat management system. Havelsan positioned Advent-AI as a response to information saturation inside combat centers where operators must process simultaneous tactical data originating from AESA radars, sonar arrays, UAV feeds, tactical data links, electronic warfare systems, satellites, and external intelligence networks. During swarm attacks, multi-axis missile engagements, dense civilian maritime traffic, or electronic warfare operations, crews face increasing difficulty distinguishing operational threats from background activity under compressed timelines.

Therefore, Advent-AI will shorten the observe-orient-decide cycle by filtering, prioritizing, and correlating incoming tactical data before operators manually process each track. This permits faster identification of irregular contacts, reduced screen-management workload, earlier recognition of engagement opportunities, and prioritization of operationally relevant tracks, leaving operators to supervise tactical evaluation during high-density combat scenarios. Track-level anomaly detection formed one of the most operationally significant functions of Advent-AI because it directly addresses the challenge of identifying hostile behavior within congested maritime environments such as the Eastern Mediterranean and Black Sea.

The AI layer continuously evaluates speed variation, course deviation, maneuver consistency, emission behavior, identity mismatches, and formation movement against expected traffic patterns and historical operational behavior. This is increasingly relevant as hostile reconnaissance assets and unmanned systems attempt to conceal themselves within civilian maritime traffic to complicate detection and delay engagement authorization. Havelsan also integrated predictive engagement-support functions to improve firing calculations, engagement sequencing, and tactical prioritization during rapidly evolving naval combat operations involving multiple simultaneous threats.

Earlier identification of irregular tracks can directly ameliorate reaction times during defense operations in naval environments with dense civilian and military traffic. Electronic warfare resilience represents another central component of the Advent-AI presentation, as military planners increasingly assume the use of such assets during combat operations. Demonstration scenarios included radar jamming, disrupted tactical pictures, intermittent sensor visibility, degraded tracking conditions, and surface-target classification under electronic interference. Advent-AI responds to that by continuously correlating sensor inputs to maintain target classification and tactical prioritization despite GPS degradation, communications denial, tactical-link disruption, radar clutter, and spoofed tracks intended to saturate operator attention.

Instead of requiring crews to manually review every incoming contact during such operations, the Advent-AI layer filters tracks according to behavioral anomalies, threat relevance, and engagement priority. The capability also aligns with NATO and U.S. Navy concerns regarding CMS survivability and decision continuity during prolonged electronic warfare campaigns against peer adversaries. For its part, the human-machine interaction layer focuses on reducing operational friction inside information centers during simultaneous air defense, surface warfare, electronic warfare, and navigation management operations. Havelsan emphasized that AI functions remain limited to recommendations, prioritization, and monitoring while operators retain full engagement authority.

Voice-assisted interaction was also introduced to reduce manual console navigation and accelerate access to tactical information during high-tempo combat operations where crews simultaneously monitor missile defense systems, tactical tracks, drone operations, electronic emissions, and navigation hazards. Advent-AI also incorporates intelligent system monitoring, possibly linked with the MAIN artificial intelligence framework introduced in 2025, which was initially focused on troubleshooting support, maintenance procedure retrieval, and operator assistance.

Additional modules monitor surrounding maritime conditions and operational risks to provide early warning concerning navigation hazards, unsafe maneuvers, and system anomalies affecting mission continuity. Advent-AI is also integrated into Türkiye’s broader distributed maritime operations concept, which links manned and unmanned naval systems through a common tactical environment. Havelsan previously integrated Advent derivatives into the Sancar unmanned surface vessel program, enabling unmanned platforms to exchange tactical information directly with crewed warships through the same command architecture.

Multiple Advent-enabled platforms can therefore function as a coordinated operational network supporting remote sensors, attritable unmanned systems, and manned-unmanned teaming concepts. For instance, the TCG Anadolu already operates Advent configurations integrated with Bayraktar TB3 operations, as part of a wider Turkish naval aviation and unmanned warfare structure. Advent-AI consequently functions not only as a shipboard decision-support layer but also as a coordination mechanism for geographically dispersed naval formations operating within network-centric warfare concepts.

The export implications of Advent-AI are significant because the baseline architecture already equips Turkish Navy vessels, Pakistan Navy MILGEM corvettes, Indonesian naval programs, Ukrainian naval initiatives, and Chilean M-class frigate modernization efforts. Havelsan indicated that the modular software structure of Advent allows AI-enabled functions to be integrated through software upgrades without replacing core combat-system hardware, reducing modernization costs and integration timelines for existing operators. This mirrors broader software-defined modernization approaches visible in systems such as Aegis Combat System, Tacticos, 9LV, and Setis, where operational capability increasingly depends on software evolution.

For smaller and medium-sized navies, the Advent-AI therefore lowers the financial barriers associated with integrating AI-assisted combat-management functions while extending operational relevance through phased modernization. Havelsan’s emphasis on sovereign AI infrastructure also supports Türkiye’s effort to reduce long-term dependence on foreign defense software ecosystems while retaining national control over combat system architecture, cybersecurity, and operational data management.

Written by Jérôme Brahy

Jérôme Brahy is a defense analyst and documentalist at Army Recognition. He specializes in naval modernization, aviation, drones, armored vehicles, and artillery, with a focus on strategic developments in the United States, China, Ukraine, Russia, Türkiye, and Belgium. His analyses go beyond the facts, providing context, identifying key actors, and explaining why defense news matters on a global scale.