Everyone loves the headline version of this: an unmanned ground vehicle that can roll up, power a laser, charge drones, and act like a roaming power plant. On paper, Hypercraft’s Razorback sounds like the kind of practical breakthrough that makes modern units faster, quieter, and harder to target.
And yet the part that makes me nervous is the same part that makes it exciting: it’s not just a vehicle. It’s a new center of gravity.
Based on what’s been shared publicly, Hypercraft unveiled the Razorback UGV, a diesel-hybrid-electric platform with 300 horsepower that can move fast and far for a ground robot—reported range around 280 miles and top speed up to 60 mph. The key detail, though, is the exported power: 38 kilowatts. That’s enough to run serious equipment in the field, including directed-energy systems, plus the more everyday but equally important jobs like charging drones and supporting forward command posts.
If you build radar and sensor-fusion systems like we do, you immediately see the shift. Power is not a side detail anymore. Power is the bottleneck. The unit that can generate and move power reliably can run more sensors, keep more drones in the air, and keep more computing online. That changes who sees first, who decides first, and who gets to hold ground without a big, obvious footprint.
But it also changes what becomes worth hunting.
A platform like Razorback pulls a lot of things toward it. Operators will want to park it near the people and systems that matter: the command post, the drone launch point, the jammer, the radar, the laser, the comms. That’s logical. It’s also a gift to the other side if they can find it.
This is where our world comes in: detection, classification, and warning. If you’re running radar drone detection around a forward position, the “threat” isn’t only the drone itself. It’s what the drone is trying to find. A cheap quadcopter doesn’t need to destroy a power platform. It just needs to spot it, mark it, and let something else do the rest.
And a power-export UGV is not a normal truck you can swap out casually. It’s a high-value node. If the unit starts treating it like a mobile wall outlet, it will end up with a queue of dependencies. “We can’t move yet, we’re charging.” “We can’t relocate the command post, we need the generator vehicle here.” “We can’t run the laser unless the Razorback is in position.” That’s how you create a single point of failure without meaning to.
I’m not saying the idea is bad. I actually think this is the direction ground forces are going, whether people like it or not. Batteries, hybrid drivetrains, electric export—those aren’t lifestyle choices. They’re survival tools when the fight is full of sensors and drones and electronic noise. A vehicle that can autonomously bring power forward could reduce risky resupply runs and keep units operating longer with less exposure.
But the execution risk is massive.
Autonomy is not a magic word. When you put an unmanned platform in the field, it becomes one more thing that can be confused, spoofed, jammed, or simply misread by tired humans under stress. Imagine a Razorback tasked to move power to a forward element at night. It arrives a little late, or takes a path that’s “safe” for it but wrong for the unit. Meanwhile the unit planned its whole sensor posture around that extra 38 kilowatts. Now your radar coverage is thinner, your drone batteries are low, and you’re choosing between visibility and silence.
Or imagine a different scenario: you’re using the Razorback to power a directed-energy system. Great—until the other side realizes that the best way to reduce that laser threat is not to outfly the beam, but to pressure the power source. Suddenly, the UGV isn’t a support asset. It’s the target that shapes the whole fight.
There’s also a quieter issue people ignore: when you make power mobile and autonomous, you encourage more gear to come forward. More sensors. More compute. More emitters. That can help you win. It can also make you louder in ways you don’t intend. Even if the Razorback itself is efficient, the systems it enables may create signatures—movement patterns, RF behavior, thermal profiles—that a smart opponent can learn. Our sensor-fusion work exists because single sensors lie. Patterns don’t.
To be fair, there’s a strong argument on the other side: centralizing power into a capable unmanned platform can simplify logistics and reduce the number of separate generators and vehicles a unit needs to protect. One rugged system might be easier to maintain than a messy mix of ad hoc solutions. And if it’s unmanned, you can accept more risk in where you send it. Those are real advantages.
Still, I think the real test is discipline. Will units treat a platform like Razorback as a flexible tool, or will they let it become the centerpiece they can’t operate without? If it becomes the centerpiece, every enemy planner will build a playbook around locating it. That means the teams running drone defense and radar drone detection won’t just be watching the sky—they’ll be defending an ecosystem that now has a very expensive heartbeat.
So here’s the question I can’t shake: are we building mobile power platforms to make units more resilient, or are we accidentally creating new “must-have” targets that make units easier to break?