US Army selects Kratos for Javelin tank killer missile seeker upgrade after Ukraine war lessons.

On August 6, 2026, the U.S. Army DEVCOM C5ISR Center awarded Kratos Defense a contract to develop, manufacture, and test a next-generation infrared seeker for the FGM-148 Javelin anti-tank guided missile system in Birmingham, Alabama. The procurement initiative targets technical obsolescence in the missile's legacy guidance section while expanding the supplier base beyond traditional prime contractors. This engineering effort ensures continuous manufacturing viability and optical tracking capability for close-combat formations through 2050.

Development and initial fabrication will take place at the Kratos Advanced Manufacturing Center prior to U.S. Army qualification testing. The replacement seeker replaces aging microelectronics architecture to support planned production expansion up to 3,960 Javelin missiles annually while maintaining target acquisition performance against armored forces.

Related topic: AeroVironment Switchblade 400 drone joins US Army LASSO program to replace traditional anti-tank missiles

The contract with Kratos aims to replace the Javelin missile’s aging target-tracking sensor with a new infrared seeker so the missile can continue finding and automatically following targets after launch for years to come. (Picture source: Lockheed Martin)

On August 6, 2026, Kratos received a contract from the U.S. Army to develop, manufacture, and test an enhanced seeker for the FGM-148 Javelin anti-tank guided missile (ATGM), with a focus on replacing aging technology in the missile's infrared guidance section. Kratos will conduct development and manufacturing at its Advanced Manufacturing Center in Birmingham, Alabama, after which this new production-ready seeker will undergo U.S. Army testing. The award also changes the Javelin's supplier structure: overall weapon development, production, and support remain under the Lockheed Martin-RTX Javelin Joint Venture, but Kratos will develop the next-generation seeker.

The contract also arrives less than three months after the delivery of the first Lightweight Command Launch Units (LWCLU) to the U.S. Army, a launcher 25% lighter and 30% smaller than the Block 1 unit, and which roughly doubles target detection and recognition range. These upgrades are occurring while the United States rebuilds an inventory heavily drawn down by the Ukraine war: Kyiv had received more than 10,000 Javelins by January 2025, compared with an estimated U.S. prewar inventory of 20,000 to 25,000 units. At the same time, Lockheed Martin is increasing its Javelin annual production capacity from roughly 2,100 missiles to 3,960 per year by late 2026, an 88.6% increase.

The U.S. Army is therefore trying to sustain a missile introduced in 1996 through at least 2050, restore inventories depleted by one high-intensity war, nearly double annual manufacturing capacity, and determine how many $200,000 guided missiles should remain when drones costing hundreds or thousands of dollars can engage many tanks at greater distances. The Kratos contract addresses the component that makes the Javelin's fire-and-forget engagement sequence possible. The Command Launch Unit (CLU) detects and identifies the target, the gunner places the aiming cursor over it and establishes lock-on-before-launch, and the missile's imaging infrared seeker then receives the target information and autonomously tracks the selected object after launch.

Unlike the TOW or the M47 Dragon that the Javelin replaced, the gunner does not have to keep a reticle on the target and transmit guidance corrections throughout missile flight, allowing the team to relocate, reload, or take cover immediately after firing. The consequence is that the seeker became one of the ATGM's critical functional dependencies: propulsion can accelerate the missile and the tandem warhead can penetrate armor only if the seeker first establishes and then maintains a valid target track. The current Javelin seeker relies on infrared detector and processing components that belong to an electronics architecture developed over decades, creating the familiar problem of semiconductor families, detector hardware, and manufacturing processes disappearing while the missile itself remains militarily useful.

Kratos' task is therefore to replace that technological dependency with new hardware suitable for continuing production, not merely to increase target range by installing a more powerful sensor. If the replacement seeker completes the U.S. Army qualification, the industrial consequence will be unusual: a subsystem currently supplied by Lockheed Martin will have been redesigned by Kratos; Raytheon and Lockheed Martin continue to co-manage the Javelin; and Raytheon remains responsible for the Command Launch Unit, missile guidance electronics and system software elsewhere in the weapon's production structure. The FGM-148 Javelin entered service with the U.S. Army in 1996 after replacing the wire-guided M47 Dragon, but the Javelin itself had already been in full-rate production since 1994.

The complete weapon weighs roughly 49 lb, with the legacy Command Launch Unit accounting for roughly 14 lb and an individual missile round roughly 35 lb, allowing a small dismounted team to carry it without needing a vehicle-mounted launcher. After launch initiation, the Javelin missile is soft-ejected from the tube before its main motor ignites, reducing the immediate exhaust effects at the firing point and allowing employment from buildings, bunkers, and other positions where conventional rocket backblast would create greater restrictions. The missile has been cited as travelling 1,000 ft every seven seconds, equivalent to 143 ft/s or 43.6 m/s, while the warhead arms after 213 ft, or 65 m, which corresponds to roughly 1.5 seconds at that nominal velocity.

Its tandem HEAT warhead uses a precursor charge against explosive reactive armor (ERA) before the primary shaped charge attacks the underlying protection, with cited penetration of 597 to 800 mm of steel. Moreover, the Javelin can employ either direct attack or an arched top-attack profile that brings the missile onto the roof of a tank rather than its heavily protected frontal arc. The seeker is consequently responsible for autonomous target discrimination during an engagement in which the operator has already ceased guiding the missile, and its reliability directly determines whether a Javelin team can exploit the tactical advantage of a weapon which passed its 50,000th missile delivery in 2021. Ukraine provides the clearest measure of both the weapon's usefulness and the danger of overstating that usefulness.

At the beginning of February 2022, Ukrainian forces possessed roughly 150 Javelin launchers and only 1,000 to 1,200 missiles, equivalent to 6.7 to 8 missiles per launcher, alongside roughly 2,000 NLAW and large quantities of other anti-armor weapons. During the opening phase, an estimate associated 300 Javelin launches with 280 Russian armored vehicles neutralized, producing a nominal 93.3% ratio. The 300 missiles alone would have represented 25 to 30% of Ukraine's pre-invasion Javelin inventory, demonstrating how quickly a tactically successful weapon can consume a relatively shallow missile stock. Even large corrections to the 280 figure would still imply substantial initial effectiveness: reducing the vehicle count by 20% gives 224 neutralized vehicles and a 74.7% ratio, reducing it by 30% gives 196 and 65.3%, and reducing it by 40% still gives 168 and 56%.

Those opening conditions were unusually favorable to portable ATGMs because Russian mechanized formations moving toward Kyiv, Chernihiv, Sumy and Kharkiv were repeatedly channelled onto roads and through settlements, bridges, forests and other restrictive terrain, while infantry screening, reconnaissance, communications and logistical support were inconsistent in several formations. The early figure therefore should not be interpreted as a scientifically established 93% probability of destroying a Russian tank. "Taken out" included armored vehicles rather than only tanks, could include immobilization or abandonment rather than catastrophic destruction, several missiles can be fired at one target, and the underlying engagement-by-engagement data were never sufficient to calculate separate probabilities for acquisition, hit, penetration, and destruction.

By January 2025, U.S. commitments to Ukraine had risen above 10,000 Javelins, 10,000 TOW missiles, and 120,000 other anti-armor weapons and munitions. By May 30, 2026, Ukraine's cumulative claim for Russian losses had reached 11,960 tanks and 24,643 other armored fighting vehicles, or 36,603 tanks and AFVs combined. On paper, 300 to 500 Javelin-neutralized vehicles would represent only 0.8 to 1.4% of claimed cumulative Russian tank and AFV losses. That percentage, however, is not an adequate measure of operational effect because Javelin engagements were concentrated disproportionately in the opening phase when Russian forces were attempting to reach Kyiv and other major objectives.

Destroying or immobilizing several hundred armored vehicles in February-April 2022 had greater operational consequence than achieving the same number of kills spread evenly across four years of positional warfare because early Russian formations were attempting rapid penetration towards critical objectives rather than fighting for incremental terrain. At the same time, the Javelin was never operating alone: Ukraine began the invasion with roughly twice as many NLAW as Javelin missiles, deployed its indigenous Stugna-P extensively, and subsequently received or produced large inventories of TOW, Stugna-P, Cobra, Kombat and other ATGMs. As the war evolved, artillery, mines, grenade-dropping UAVs and FPV drones increasingly absorbed the armored target set, particularly after persistent drone reconnaissance made it more dangerous for ATGM teams to approach and remain within several kilometers of Russian positions.

Russian vehicle protection changed accordingly: early turret cages, intended to complicate top attacks, were followed by much larger roof structures, mesh, chains, additional ERA, and electronic warfare equipment on so-called turtle tanks, primarily in response to continuous FPV drone exposure rather than specifically to the Javelin. The Javelin's declining share of kills is therefore better understood as a change in the battlefield's weapons mix than evidence that the missile suddenly lost its ability to attack tanks. The problem for future U.S. procurement is therefore the cost and depth of that weapons mix. At a cited Javelin missile price of $200,348, replacing 10,000 rounds at the same unit benchmark would cost $2.00348 billion for missiles alone, before Command Launch Units (CLUs), batteries, training equipment, spare parts, transportation, storage and sustainment.

A $500 FPV represents 1/401 of that missile price, meaning the nominal procurement cost of one Javelin corresponds to about 400 such drones. Even using a $5,000 FPV drone, one Javelin corresponds financially to 40 drones; five Javelins cost roughly $1.002 million, equivalent to about 200 FPVs. Another comparison using a $2,000 drone figure produces a ratio close to 100:1. If five $2,000 FPVs are required to disable a tank, the munition expenditure is $10,000; even ten drones would cost $20,000, one-tenth of a single Javelin. Engagement depth reinforces that cost ratio. The Javelin's maximum envelope is roughly 4 to 4.75 km, while Ukrainian FPV attacks can exceed 10 miles, or 16 km, meaning a drone can now have more than three times the Javelin's reach and can search behind hills, buildings, or tree lines without the launching operator initially seeing the target. However, the comparison is not one-for-one.

FPVs depend on radio links and operators, are vulnerable to jamming, weather, as well as interception, carry smaller warheads, and can require several attacks against a tank, whereas the Javelin carries a purpose-built tandem HEAT warhead and no longer depends on a control link once its infrared seeker has locked and the missile has left the tube. The procurement implication is therefore not that FPVs replace the Javelin (or even a tank), but that a force firing $200,000 missiles against every vehicle or position will consume high-end inventory far faster than a force reserving Javelins for heavily armored, time-sensitive, or electronically contested targets and using cheaper drones elsewhere. Moreover, the 2026 LWCLU partially changes the Javelin's cost-effectiveness because the reusable launcher has a broader military utility before a missile is expended.

The first units were delivered to the U.S. Army on May 26, 2026, after a contract was awarded in June 2022. The new unit is 25% lighter and 30% smaller than the Block 1 CLU and doubles day-and-night target detection and recognition range. Starting from a legacy CLU weight of roughly 14 lb, a 25% reduction removes about 3.5 lb, leaving an implied launcher weight near 10.5 lb and reducing the complete missile-plus-launcher burden from roughly 49 lb to about 45.5 lb. More important, doubling detection and recognition range creates an observation envelope larger than the missile's firing envelope, allowing the Javelin team to locate and classify targets without automatically committing a missile. This matters because the CLU has always been removable from the missile and usable as an independent thermal observation device; the LWCLU increases that surveillance function while reducing weight and volume.

The launcher is compatible with existing and future Javelin missiles, so the U.S. Army can introduce the new sight without discarding older missile lots or maintaining launcher-to-missile compatibility tables. Therefore, Raytheon invested $22 million in its Tucson, Arizona, LWCLU facility to increase production speed and capacity, while the list of users operating or procuring the lighter launcher already includes Australia, Estonia, Latvia, Lithuania, Poland, the United Kingdom and the United States. Modernization is consequently occurring at both ends of the engagement chain: the LWCLU increases the range at which the operator can detect and recognize a prospective target, while the Kratos seeker is intended to keep the Javelin's fire-and-forget mode viable as legacy guidance components become obsolete.

Like the Patriot, industrial capacity remains the limiting factor if the United States expects to simultaneously rebuild inventories and supply foreign customers. The Javelin's pre-expansion production rate was roughly 2,100 missiles per year, or 175 per month; a Pentagon-reported increase put actual output near 2,400 per year, or 200 per month, while Lockheed Martin's late-2026 capacity target is 3,960 annually, or 330 per month. The planned increase from 2,100 to 3,960 missiles represents an 88.6% rise in capacity, but replacing 10,000 missiles would still require 2.53 years at 3,960 per year if the entire production capacity were dedicated to replacing those rounds and none were allocated to new U.S. requirements or foreign customers. At the former 2,100-per-year rate, the same replacement requirement would take 4.76 years.

Increasing final assembly cannot solve the problem by itself because the Javelin supply chain contains nearly 100 part-level suppliers and 25 major subcontractors producing items from propulsion components to guidance electronics. Those suppliers required 8 to 10 months of non-recurring engineering work to install additional tooling, test equipment, and manufacturing space before they could support increased output, while the program is working to keep critical component lead times below 52 weeks, qualify alternative sources for high-risk items, and accumulate reserves of critical materials. Propulsion manufacturing offers one example of potential cycle compression: Firehawk Aerospace reduced the solid-propellant manufacturing cycle from a conventional 15 to 60 days of curing to 3 to 6 hours using additive manufacturing, equal to a 120-fold reduction when comparing 15 days with three hours and a 240-fold reduction when comparing 60 days with six hours.

That does not make a complete missile producible in hours because motor assembly, seeker manufacture, warhead production, electronics integration, inspection and acceptance testing remain separate bottlenecks, but it shows why the Javelin production capacity must be measured across the whole industrial chain rather than only at final assembly. If Javelin remains operational through at least 2050, the tank-killer missile will span roughly 54 years from its 1996 service entry and 56 years from the start of full-rate production in 1994, a life cycle long enough for several generations of electronics and manufacturing technology to disappear while the military requirement remains. The Kratos seeker contract is therefore best understood as one element of a larger effort to prevent component obsolescence from terminating production of a weapon whose basic combat architecture remains useful.

By 2026, that effort combines a production-ready replacement seeker under development in Birmingham, a launcher delivered from Tucson that cuts weight by 25%, dimensions by 30% and doubles detection and recognition range, a $22 million launcher production investment, an 88.6% planned increase in annual missile capacity, from roughly 2,100 to 3,960 units, supplier expansion across roughly 125 major supplier and subcontractor relationships, subcomponent lead-time targets below 52 weeks and propulsion processes being tested with production cycles measured in hours rather than weeks. The operational requirement, however, is changing faster than the missile itself.

Ukraine consumed Javelins at a rate sufficient for 300 missiles to equal 25-30% of its pre-invasion stock in the opening phase, eventually received more than 10,000 U.S. missiles, and then moved into a battlefield where $500 to $5,000 FPVs could attack many of the same vehicles from distances exceeding 10 miles. However, the Javelin remains relevant where infantry require an immediate autonomous engagement against armor, where electronic warfare makes radio-controlled drones unreliable, where a target must be attacked within seconds rather than searched for over several minutes, or where a tandem HEAT warhead is required to defeat Russian tanks.

The U.S. force planning issue through 2050 is consequently a stockpile-sizing problem: 10,000 missiles represent only 2.5 years of production at the planned 3,960-per-year capacity, but also more than $2 billion at a $200,348 missile price, while the same procurement expenditure could finance hundreds of thousands of low-cost FPVs depending on configuration. The objective is therefore not to maximize Javelin numbers independently of the wider force, but to maintain enough missiles for the subset of anti-armor engagements where their autonomous seeker, rapid reaction time, portability and tandem warhead provide capabilities that cheaper drones, mines, artillery or other munitions cannot provide with comparable reliability.

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.

Explore More Defense News

• Land Defense News
• Naval Defense News
• Defense Aerospace News