Because the drone {industry} pushes towards higher autonomy, scalability, and integration into shared airspace, its dependence on superior semiconductors is changing into unattainable to disregard. On this visitor put up, Michelle Duquette of 3 MAD AIR explores why nearer collaboration between the semiconductor and aviation communities is crucial to handle certification, provide chain resilience, and long-term operational security. Her perspective highlights how aligning chip design with aviation necessities right this moment might help keep away from expensive obstacles to drone deployment tomorrow. DRONELIFE doesn’t make or settle for cost for visitor posts.
When Chips Take Flight: Constructing the Semiconductor-Drone Partnership
by Michelle Duquette, Founder and CEO of 3 MAD AIR Consulting.
Final week, I had the privilege of becoming a member of a panel on the Northeast Semiconductor Manufacturing Hall Workshop in New York Metropolis-a gathering that introduced collectively semiconductor producers, quantum computing corporations, and photonics consultants from throughout the united statesNortheast and Canada. Canadian companies,supported by the Canadian Consulate, have been notably energetic within the discussions. It was obvious that North American semiconductor and aerospace collaboration doesn’t cease on the border. I used to be the one aerospace skilled within the room, and I went particularly to open a dialog between two sectors that desperately want one another however hardly ever speak.
Panel “Built-in Circuit Design from a Protection Business Perspective.”
The panel, titled “Built-in Circuit Design from a Protection Business Perspective,” introduced collectively quantum computing innovators and AI chip designers, and me, an aviation strategist who spends her days serving to operators, regulators, and expertise suppliers make remotely piloted and autonomous plane function safely in shared airspace. What turned clear over the course of the dialog is that drone aviation and semiconductor manufacturing are two industries that critically want one another, and that the challenges we’re going through are interconnected.
The Foundational Dependency
Let me be direct: the drone {industry} can not scale with out trusted, aviation-grade semiconductors. Each functionality we’re attempting to allow, from autonomous flight controls, AI-powered decision-making, safe communication, or detect-and-avoid relies upon fully on chips that may carry out below strict Measurement, Weight, Energy, and Value (SWaP-C) constraints whereas assembly aviation certification requirements. That is about constructing a really interoperable Nationwide Airspace System from the bottom up.
And the timing couldn’t be extra essential. The FAA’s Model New Air Visitors Management System (BNATCS)-a $12.5 billion effort to rebuild America’s air visitors infrastructure by 2028-is being architected proper now. If the semiconductor and aviation communities don’t align right this moment, we danger constructing a system that’s optimized for legacy manned plane however can’t natively deal with the heterogeneous visitors mixture of the 2030s. We’d like unified information architectures and open requirements baked into BNATCS Section 1, not bolted on later as an afterthought.
And right here’s what the semiconductor {industry} wants to listen to from us within the aviation {industry}: you may’t simply take the most recent gaming GPU off the shelf and name it flight-ready. Aviation calls for provenance, secure-by-design architectures, long-term lifecycle help, and chips that gained’t be obsoleted on shopper product timelines. When a public security drone fleet is grounded as a result of a essential radio chip reached end-of-life with no various supply, it surpasses inconvenience and instantly turns into a nationwide safety vulnerability. Drones deployed for good, like public security, catastrophe response, and demanding infrastructure inspections can’t afford to fail due to provide chain fragility.
Widespread Issues, Unusual Collaboration
Because the panel unfolded, I saved listening to echoes of my very own interoperability challenges mirrored in what the semiconductor people have been describing.
I see the identical sorts of gaps day-after-day in my work. I’m serving to states construct AAM sandboxes that want trusted semiconductor provide chains. I’m working with operators and repair suppliers who can’t discover home aviation-grade chips with predictable lifecycles. I’m connecting expertise suppliers who don’t even notice the opposite exists. And I’m watching the FAA spend numerous hours checking out security instances for plane whose chip designs don’t fairly meet mission necessities, leading to extra danger assessments, lists of compensating mitigations, and certification-related one-offs that might have been prevented if the semiconductor and aviation communities had been speaking early on.
That regulatory burden isn’t simply an FAA drawback. It actually faucets the brakes industry-wide on innovation. Each time a chip is designed with out understanding aviation constraints, or an plane integrates elements with out anticipating nationwide safety or certification necessities, the FAA has to untangle the implications. Is the processing latency acceptable for detect-and-avoid? What occurs when the chip reaches end-of-life mid-certification? How do you validate an AI accelerator’s conduct below all flight situations? That is our every day actuality slowing down approvals and driving up prices for everybody.
• Provide chain fragility: Each industries rely upon layered infrastructure-raw supplies, fabrication, packaging, testing, integration. A weak spot wherever in that chain breaks the entire system. In semiconductors, it’s rare-earth provide and fab capability. In drones, it’s aviation-grade elements that may’t be single-sourced from adversarial suppliers.
• The valley of demise: Promising prototypes that by no means grow to be licensed, deployable merchandise. Semiconductor corporations battle to navigate protection acquisition processes. Drone operators battle to maneuver from one-off approvals to scalable, repeatable operations. Each industries are caught in bespoke validation cycles which might be too sluggish and too costly.
• Export controls and dual-use complexity: Excessive-performance chips, particularly AI accelerators, fall below ITAR and EAR restrictions, identical to superior drone radios, sensors, and flight controllers. For those who don’t design for compliance from day one, you danger constructing methods that may’t be deployed internationally or shared with key companions.
• Protection-adjacent vs. defense-only growth: Do you construct separate methods for industrial and army purposes, or design one trusted platform that serves each? The semiconductor {industry} already is aware of the reply: they don’t construct one-off prototypes for a single buyer. They develop scalable, validated processes that serve a number of markets. That’s precisely the mannequin we want in aviation.
Precision Manufacturing Meets Precision Operations
One analogy saved surfacing throughout the dialogue: each industries function in zero-margin-for-error environments.
In a chip fab, a single contamination occasion or course of deviation can destroy a complete wafer batch. In aviation, a single counterfeit element or meeting error can floor a fleet or trigger catastrophic failure. Each require excessive precision, cleanroom-level course of self-discipline, rigorous traceability, and batch testing. The Northeast’s semiconductor heritage of precision manufacturing, protection electronics, and supplies science maps instantly onto the necessities for constructing trusted, certifiable drones.
Sadly, right this moment we’re optimizing these constraints in isolation, and that’s what breaks us. Semiconductor corporations optimize for efficiency per watt with out understanding mission profiles. AI groups optimize for compute with out understanding drone flight dynamics. Plane integrators optimize for weight and energy with out understanding what the chips truly have to ship the mission or the conclusion they’ll affect these designs. So we find yourself with chips which might be too power-hungry for the airframe, AI accelerators that demand computing energy the platform can’t help, and flight controllers that may’t deal with the processing latency necessities for detect-and-avoid.
SWaP-C isn’t nearly making a chip smaller or extra environment friendly. It’s about understanding that flight controllers, sensors, radios, batteries, and AI accelerators all compete for a similar energy envelope and bodily house on an airframe, and that each resolution in a single area constrains what’s attainable in one other.
What Actual Collaboration Seems Like
From my view, the answer is co-design from day one.
Get chip designers, AI engineers, drone integrators, and operators in the identical room. Map the mission necessities first, then work backward to the SWaP-C tradeoffs. That’s the way you keep away from over-speccing a system that may’t fly or under-delivering a functionality that doesn’t meet the mission. Actual collaboration means constructing strategic partnerships the place everybody shares a standard imaginative and prescient of success. It’s interoperability.
Through the panel, we talked about what this might appear to be in apply for the Northeast corridor-and the broader North American ecosystem:
• Shared sandboxes and qualification pipelines the place AI {hardware} and software program get validated collectively below actual aviation situations and never in separate silos.
• Co-located growth the place semiconductor designers, AI labs, drone and eVTOL producers, operators, and regulators iterate on certification, security instances, and mission efficiency in actual time.
• Cross-border consortia the place chipmakers, integrators, and aviators collectively map essential elements and establish single factors of failure earlier than a disaster hits.
• Twin-use check environments that function “fabs of the drone ecosystem”: shared infrastructure that reduces obstacles to entry, accelerates validation, and allows small corporations to compete.
Beginning with Actual Missions
Right here’s my ask to the semiconductor neighborhood: Don’t look ahead to the right summary use case.
Choose an actual mission, like a North Nation winter storm response, maritime search and rescue off the coast of Maine, essential infrastructure inspection in Quebec, and construct the stack collectively. Show that AI, semiconductors, and aviation can ship tangible public profit and dual-use worth in an operational setting with actual accountability.
These are precise deployments with actual plane operators, actual airspace constraints, and actual penalties if the expertise doesn’t work as marketed. That is operational pragmatism at its core, specializing in confirmed expertise that solves actual issues.
And whenever you do that, apply the identical self-discipline that will get you from fab to flight-ready: precision at each layer, traceability, redundancy, and proactive danger administration. The aviation neighborhood is aware of how costly it’s to retrofit resilience or interoperability. Design for it from the beginning.
An Invitation
The Northeast Semiconductor Manufacturing Hall has a once-in-a-generation alternative to be the place the place two industries that desperately want one another cease speaking previous each other and begin constructing collectively.
The problem: semiconductor corporations don’t know which aviation rules apply to their merchandise or tips on how to map their roadmap to certification pathways. Drone operators can’t articulate SWaP-C necessities in language chip designers perceive. States and universities constructing check sandboxes want either side on the desk however don’t all the time know tips on how to convene them successfully. And everyone seems to be burning money and time on integration makes an attempt that drag on or fail as a result of nobody is translating between domains.
That is precisely the bridge-building work I do. I spent 24 years contained in the FAA’s technical structure, and now I work every day with the drone and eVTOL neighborhood scaling actual deployments, states standing up interoperable drone and AAM ecosystems, and expertise suppliers attempting to navigate certification and provide chain resilience. I translate between regulators, operators, and expertise suppliers so that you don’t need to determine it out alone and so the FAA doesn’t need to untangle preventable dangers after the very fact.
For those who’re a semiconductor firm attempting to know what “aviation-grade” actually means for the drone world, or a drone operator pissed off by chip suppliers who don’t perceive your constraints, or a policymaker questioning tips on how to speed up each industries, let’s come to the desk.
The issues are laborious, however they’re solvable. And none of us can clear up them alone.
As a result of on the finish of the day, chips don’t fly themselves. And drones don’t work with out the fitting chips.
It’s time we figured this out collectively.
Michelle Duquette is CEO of 3 MAD Air, a consulting agency specializing in drone and superior air mobility interoperability. With 35 years in aviation-including 24 years at MITRE Company leading to main the FAA’s UAS and AAM analysis portfolio-she helps bridge the gaps between operators, regulators, expertise suppliers, and policymakers to allow secure, scalable operations in shared airspace.
Learn extra:
