This is exactly the kind of counter-drone system that looks “obviously needed” right up until you ask one uncomfortable question: what happens when everyone starts putting 40 mm guns on trucks and calling it air defense?
Public reporting says Leonardo shared the first photo of a Hystrix 40 ADS prototype. It’s a truck-mounted 40 mm anti-aircraft gun built on an Iveco Astra 8x8 chassis. They’ve taken their naval Marlin 40 ILOS turret and adapted it for land, then added an electro‑optical setup meant to find and track small drones. The whole thing is positioned as short-range air defense, with a very clear focus on low, small targets.
From a sensor company point of view, the most interesting part isn’t the gun. It’s the targeting stack.
Hystrix leans on an EO station called Janus D: thermal imaging, a digital TV camera, and a laser rangefinder. That’s a solid package for identifying something small that doesn’t want to be seen. And it matches what we’re watching in real life: drones aren’t just “flying objects” anymore. They’re small, cheap, fast to replace, and often flown in messy environments where you’ve got birds, wires, dust, smoke, and random heat sources. EO can see a lot, but it can also get fooled, blocked, or overwhelmed.
That’s why our bias is simple: EO-only solutions are fragile. They can be great when conditions are good and the operator has time. But the reason “radar drone detection” keeps coming up is because radar doesn’t care if it’s dark, hazy, or the drone is painted a clever color. The hard part is not “can you see a drone,” it’s “can you keep seeing it long enough to make a good decision,” and can you do that while ten other things are happening.
Leonardo is also pairing this with programmable ammunition. The reporting mentions several ammo types, including programmable-fuze rounds and multifunction programmable rounds with multiple detonation modes. That’s not a small detail. It’s a signal that the industry has accepted something uncomfortable: hitting a small drone with a single direct shot is not a plan. It’s a hope. Airburst and programmable effects are the realistic answer when targets are small and moving and you don’t get a perfect track.
They also shared claimed engagement ranges: small UAVs at 1–1.5 km, subsonic missiles up to 1.6 km, and helicopters/fighters up to 3 km. If those numbers hold up in real conditions, it’s a serious short-range option. But that “if” is doing a lot of work. Range on paper is not the same as range under jamming, bad weather, cluttered terrain, and a target that is actively trying to confuse you.
Here’s the tension I don’t want people to ignore: the gun is the easy part. The decision loop is the hard part.
Imagine you’re protecting a power substation on the edge of a city. You get a track on something small. Is it a quadcopter filming? A one-way attack drone? A bird? A friendly drone that drifted? If you’re wrong and you fire, you don’t just “miss.” You send 40 mm ammunition into a place where people live and work. The more effective the gun is, the more serious the rules for using it need to be.
Or imagine you’re defending a convoy. One drone appears, then two, then five. The turret can fire fast, and it reportedly holds 72 rounds. That sounds like plenty until it isn’t. This is where sensor fusion and prioritization stop being buzzwords and become survival. If your system can’t fuse radar drone detection with EO and other inputs to keep tracks stable and classify targets quickly, you either waste ammo on junk or you hesitate until it’s too late.
Now, a fair counterpoint: there’s a reason companies lean into EO. EO helps with identification. It gives operators confidence. It can reduce bad shoots. And in some environments, radar is a magnet for interference and deception. That’s real. But we don’t think the answer is choosing one sensor type like it’s a religion. The answer is redundancy and fusion, because the enemy gets a vote and the environment gets a vote.
One more detail matters: this turret can reportedly be removed from the vehicle and deployed as a stationary installation. That’s practical, and it hints at how these systems will actually get used: parked near key sites, moved when needed, integrated into layered defense. That layered part is where things either work or fall apart. A standalone gun with its own EO can be “good.” A gun that’s plugged into a wider sensor picture—shared tracks, fused IDs, coordinated fire control—can be decisive. But if everyone builds their own closed system and refuses to integrate, you get a bunch of expensive islands that don’t help each other.
My judgment is that Hystrix is pointing in the right direction on effectors: programmable ammo, high rate of fire, optimized for small UAVs. Where I’m skeptical is whether the industry is taking the sensor problem seriously enough. The drone threat is scaling faster than training and doctrine, and faster than the public is ready to accept the risks of firing artillery in and around real infrastructure.
If these systems spread, what standard should decide when a track is “certain enough” to justify firing in a world where the line between a toy drone, a friendly drone, and a weapon can be one software setting?