M2M Tech
For DRDC & Royal Canadian Air Force teams

MEA for Air Force Readiness

Edge AI for perimeter awareness, predictive maintenance, PPE safety, and legacy equipment digitization — processed locally, on base, with no dependency on cloud connectivity.

Proposed Pilot Use Cases for DRDC / Air Force TestingWhy MEA
Edge
Local inference
8–12 wk
Pilot window
Low / no
Cloud required
Rugged MEA Edge AI device connected to perimeter cameras, an analog control panel, generator, and command dashboard
MEA · Edge AI Node
02 · Foundation

Why MEA for Air Force environments

Air bases and support environments combine modern systems with legacy equipment. Many operational risks remain visible through cameras, gauges, lights, meters, panels, and human activity patterns. MEA converts these visual and sensor signals into structured digital events — without replacing what already works.

Local edge processing
Low-latency alerts
Works with existing cameras & legacy equipment
Supports low / no-cloud environments
Ingests camera, analog, digital & sensor data
Designed for pilot testing & operational validation
Use case · 01

Airfield & Base Perimeter Scanning

MEA monitors fence lines, restricted zones, hangars, ramps, fuel areas, and remote perimeters using fixed cameras, mobile cameras, or drone feeds.

Detects
  • People near restricted zones
  • Animals near airfield perimeter
  • Unauthorized vehicles
  • Fence-line movement
  • After-hours activity
  • Zone breaches
Operational value · Improves situational awareness, reduces manual monitoring burden, and supports faster response.
Airfield & Base Perimeter Scanning
Use case · 02

Predictive Maintenance for Ground Support Equipment

MEA connects to cameras, sensors, and analog/digital ingress modules to monitor generators, compressors, pumps, HVAC, fuel equipment, power systems, and ground support assets.

Detects
  • Abnormal readings
  • Equipment degradation
  • Runtime anomalies
  • Warning lights
  • Temperature / pressure / voltage changes
  • Maintenance risk patterns
Operational value · Improves readiness, reduces unplanned downtime, and supports condition-based maintenance.
Predictive Maintenance for Ground Support Equipment
Use case · 03

PPE and Safety Compliance

MEA monitors maintenance zones, hangars, fuel areas, workshops, and flight-line support areas for PPE and safety conditions.

Detects
  • Missing high-visibility vest
  • Missing helmet or safety glasses
  • Restricted-zone entry
  • Unsafe proximity to equipment
  • Person-down events
  • Unusual activity in controlled areas
Operational value · Supports safety compliance, training, incident review, and auditability.
PPE and Safety Compliance
Use case · 04 · Standout

Bring 1980s equipment into a modern predictive maintenance network.

Many defence environments still rely on reliable but older generators, meters, pumps, gauges, and control panels. MEA visually reads analog dials, needle gauges, seven-segment displays, indicator lights, and switch positions using camera-based AI — a non-invasive path to digitize legacy equipment without replacing or rewiring it.

Analog control panel with AI overlays detecting voltage, needle angles and warning light states
Voltage reading detectedCurrent reading detectedNeedle angle → digital valueGreen / yellow / red light stateMaintenance alert generated
Example

“If a generator's oil pressure needle drops or a warning light changes state, MEA detects the change and sends an alert to the maintenance team before the equipment fails.”

Operational value
  • No invasive integration
  • No replacement of legacy equipment
  • Rapid pilot deployment
  • Visual indicators → structured digital data
  • Predictive maintenance for old assets
  • Supports digital modernization
03 · System

MEA architecture

A simple, auditable pipeline from physical signal to operational action.

Inputs
  • · Legacy equipment
  • · Cameras
  • · Sensors
  • · Analog / digital I/O
MEA Edge AI Device
  • · Local inference
  • · Event detection
  • · On-device storage
Events & Alerts
  • · Event log
  • · Dashboard
  • · Real-time alerts
Consumers
  • · Maintenance team
  • · Command centre
  • · Digital twin
Event data outputs
Event type
Timestamp
Confidence score
Detected reading
Equipment ID
Location
Image snapshot
Recommended action
04 · Pilot approach

Suggested 8–12 week pilot

Phase 1
Select 2–3 test environments
  • Perimeter camera
  • Equipment room / generator
  • Maintenance area
Phase 2
Deploy MEA and camera / sensor interfaces
  • Connect to existing cameras where possible
  • Add temporary cameras if needed
  • Configure analog / digital ingress
Phase 3
Train and tune detection models
  • Perimeter events
  • PPE events
  • Analog gauge readings
  • Warning light states
Phase 4
Validate operational usefulness
  • Alert accuracy
  • False positives
  • Latency
  • Maintenance usefulness
  • Operator feedback
05 · Measurement

Success metrics

01Detection accuracy
02False alert rate
03Alert latency
04Legacy indicators digitized
05Safety events detected
06Equipment readings captured
07Maintenance alerts generated
08Operator feedback
09Readiness improvement potential
06 · Next step

Modernize what already exists.

MEA helps Air Force and DRDC teams test practical Edge AI use cases without replacing existing infrastructure. By turning cameras, sensors, gauges, displays, and equipment activity into actionable digital events, MEA can support safer bases, more reliable equipment, and faster operational decisions.

Proposed next step

Identify one perimeter, one equipment asset, and one safety zone for pilot testing.