Early proof,
kept honest.
The credibility work here is simple: show the direction, show the doctrine relevance, and avoid pretending the product is further along than it is.
Submitted to NATO Innovation Continuum 2026 as part of the current Maple Shield product direction.
Positioned for passive sensing close to the sensor, without active RF emissions and without assuming a rich backend.
Current metrics are still under validation. Evaluation conversations focus on deployment fit, density, and operational relevance rather than overstated maturity.
System performance
current targets.
Grouped the way an evaluator reads them: detection, performance, deployment, and reliability.
- 300-800 m for Class I UAS
- Up to 1,500 m target for Class II
- <50 ms end-to-end latency target
- 30+ FPS on Jetson-class edge GPU hardware
- <$1,000 CAD target cost per node
- <25W power target per node
- <2% false positives with multi-frame validation
- -40C to +50C target operating range with enclosure
These values are based on current prototype targets and ongoing validation. The goal is to keep claims grounded in measurable system behavior rather than theoretical capability.
CAIRN Detection
Engine
CAIRN combines real-time vision inference, multi-object tracking, and probabilistic risk scoring to identify and prioritize aerial threats under constrained edge environments. It is accelerated by the SparseFlow inference stack, enabling real-time performance on low-power edge hardware.
Real-time vision inference tuned for passive drone and sUAS sensing on constrained edge hardware.
Multi-object drone and sUAS track continuity that preserves identity through clutter, motion, and partial occlusion.
Trajectory and movement interpretation for earlier recognition of approach vectors and threat behavior.
Probabilistic scoring that prioritizes persistence, directionality, and operationally relevant behavior.
Replayable event evidence for operator review, debugging, and post-incident analysis.
Why Maple Shield wins
on deployment scale.
Maple Shield enables significantly higher sensor density at a fraction of the cost of traditional systems. The advantage is not just detecting drones. It is distributing more passive nodes across more space, with lower cost, lower infrastructure burden, and a cleaner path through fixed-camera deployments.
Low-cost passive nodes make it practical to field more sensing coverage than radar-heavy architectures and extend awareness across larger sites.
No active RF emissions keeps Maple Shield aligned with EMCON-sensitive deployments where awareness matters but unnecessary signal exposure is unacceptable.
Designed around distributed fixed-camera infrastructure so remote, Arctic, and critical-site coverage can scale without radar-heavy node economics.
Positioned against system
categories, not hype.
This is the category-level positioning Maple Shield is built around. The strongest advantage is deployment density: more passive nodes, distributed faster, at lower node cost.
| System | Cost per node | Emissions | Arctic capability | EMCON compatibility | Deployment speed | Deployment density |
|---|---|---|---|---|---|---|
| Maple Shield | <$1,000 CAD target | None | Targeted | Native | Fast | High |
| Radar-heavy systems | High | Active | Program dependent | Limited | Slower | Low |
| RF-only systems | Medium | Passive | Environment dependent | Partial | Fast | Medium |
Illustrative positioning only. Actual program fit depends on mission profile, terrain, and the wider sensor stack.
Operational scenarios that
make the value concrete.
Structured like operational notes instead of generic airspace language, so reviewers can see how the system would actually be evaluated.
- Detection: 600 m quadcopter target
- Latency: <50 ms target
- RF emissions: none
- Outcome: early warning without detection exposure
- Detection: Class I UAS within current target range
- Power: <25W per node target
- Deployment: passive edge node on commercial hardware
- Outcome: persistent awareness where logistics are thin
- Coverage: multi-node low-altitude monitoring
- Cost: <$1,000 CAD per node target
- Deployment density: significantly higher than traditional systems
- Outcome: faster site coverage and cleaner operator escalation
Who Maple Shield
is for.
This is not a mass-market drone app. The product direction is shaped around operators and programs that care about passive sensing, auditability, and low-infrastructure deployment.
Programs evaluating passive drone detection, airspace awareness, and operator-ready incident intelligence for sensitive environments.
Energy, industrial, and other site operators that need persistent awareness and cleaner escalation around low-altitude aerial activity.
Operators covering remote, border, maritime, or cold-weather locations where passive sensing and edge deployment matter more than cloud convenience.
Maple Shield should read like a serious passive-detection system.
The goal is not to oversell maturity. It is to make the category, doctrine fit, and deployment advantage clear enough that the right reviewer immediately understands why Maple Shield exists.
CAIRN is accelerated by the SparseFlow inference stack, enabling real-time performance on low-power edge hardware while Maple Shield stays focused on passive drone detection and operator-ready airspace awareness.