US Military's Shift to Low-Cost Swarms Mirrors EV Industry Cost-Efficiency Rebalancing

From "Reaper" to Swarms: An Engineer's Perspective on Cost-Effectiveness Rebalancing

The US Department of Defense has launched a tender for cheap drone swarms, attempting to replace some MQ-9A "Reaper" missions with "large numbers of expendable, cheap drones." This news blew up in military circles, but as an engineer who constantly battles vehicle costs, battery life, and electric drive system reliability, I see not "disruption," but the most familiar inflection point on the cost-effectiveness curve in engineering.

Two Paths, Two Engineering Philosophies

Path A: MQ-9A "Reaper" — "High Reliability, Long Endurance, Expensive"

Unit price is $30 million; add ground stations and training, and the full system easily exceeds $100 million. It carries heavy sensors, satellite communications, and multiple precision-guided munitions, cruising at 15,000 meters with loiter time exceeding 24 hours. But its weakness is: extremely high single-point failure risk. One shootdown doesn't just mean losing $30 million; it also breaks the intelligence chain, zeroes out trained personnel, and causes political trouble. It's like a high-end EV—100kWh battery pack, 800km range, but once the BMS (Battery Management System) has a single-point fault, the whole car stalls, and repair costs approach replacement costs.

Path B: Cheap Drone Swarms — "Low Reliability, Short Endurance, Cheap"

The tender requires unit costs squeezed down to hundreds of thousands of dollars or less, potentially using mature commercial components (like fuel engines, DJI-level flight controllers), deploying hundreds at a time. Each carries only one camera or one small bomb, but mission failure losses are acceptable. This is like our low-cost EVs: using cylindrical cells (like simplified 18650s or Blade Batteries), not pursuing single-cell energy density, but relying on large-scale series-parallel connections and thermal management design for redundancy. Even if one cell undergoes thermal runaway, the system has redundancy, and the car can still run at low power to a repair shop.

Three Key Dimensions of Engineering Feasibility

1. Endurance and Mission Radius

The MQ-9A can fly for 24 hours due to its heavy turboprop engine + large wingspan aerodynamic design. Switching to cheap swarms, they'll likely use small-displacement commercial piston engines or electric motors, with max endurance of 2-4 hours. This means they can't perform "sunbathing" continuous ISR (Intelligence, Surveillance, Reconnaissance), only "short-duration strikes" or "saturation attacks." In automotive terms, this is pure-electric urban logistics vans vs. hybrid long-haul trucks—the former charges fast and is cheap, but can't handle long highway trips.

2. Communication and Data Link Bottlenecks

The biggest engineering hurdle for swarms isn't building cheap planes, but making hundreds fly simultaneously without colliding while transmitting data in real-time. The US previously tried the "Gremlins" project and finally found that ad-hoc network data link bandwidth and latency couldn't hold up. Analogous to V2X (Vehicle-to-Everything)—we tested fleets of thousands of electric logistics vehicles; in areas with poor 5G coverage, packet loss rates spiked to 40%, forcing us to simplify communication protocols (transmitting only position and speed, not video). Similarly, if swarms transmit HD video, bandwidth explodes instantly; engineering-wise, it can only compress to "low-res snapshots + metadata transmission after AI recognition."

3. Expendability vs. Maintainability

The "Reaper" has a design life of tens of thousands of hours, requiring deep maintenance after each mission. Swarm drones might have a design life of only dozens of flights, or even "use once and discard." This drastically lowers maintenance costs but brings new engineering issues: How do you amortize the costs of catapults, recovery nets, and aerial refueling infrastructure? If every drone needs a dedicated catapult for launch, the total system cost goes up again. A more pragmatic approach is vertical takeoff and landing (VTOL) like drone delivery, but that sacrifices range.

My Judgment: It Might Succeed, But Won't Be a "Replacement"

This US military tender essentially decomposes the "Reaper's" tasks:

• Continuous surveillance and high-value target strikes continue to use the "Reaper."
• Rapid reconnaissance, air defense suppression, and saturation attrition use swarms.

This mirrors exactly the logic behind BYD's Blade Battery:

• High-end models use high-energy-density ternary lithium + CTP (Cell-to-Pack) for 800km range.
• Economy models use LFP Blade batteries + low-cost packs for 400km range, but are safe, cheap, and replaceable.

Therefore, the "Reaper" won't be replaced, but "dimensionally reduced and supplemented." Swarms will eat part of the "Reaper's" tasks but won't retire the "Reaper." Because in engineering, a single system can never cover all scenarios—unless you're willing to pay infinite costs.

An Open Question

If the US military really squeezes swarm unit prices down to $500,000, will their communication protocols and ad-hoc networking algorithms increasingly converge with civilian drone tech paths? In other words, the DJI Air 3S you buy might contain the underlying communication protocols of US military swarms in its flight controller—sounds crazy, but the auto industry has already done this (many of our automotive-grade chips are downgraded military-grade ones). So, will the "cost-function" boundary of the civilian drone market inversely reshape the degrees of freedom in military drone engineering?

---

Original Link: https://www.ithome.com/0/975/550.htm