IRISIRISIntelligent Real-time Instability Sensing
AI-enabled mine subsidence monitoring

IRIS Mine Stability Command Center

Intelligent Real-time Instability Sensing fuses wireless surface sensors, satellite deformation data, and physics-informed prediction into one operational picture of a mine's stability.

IRIS eye badge

Safer mines · Stable tomorrow

Live sensor fusion

Tilt, vibration, and displacement from every node are fused into one evidence score, so no single cheap signal can veto an escalation on its own.

Physics-informed prediction

A graph neural network grounded in angle-of-draw physics estimates trend and time-to-threshold as a range, never a false-precision date.

Digital twin

A 3D terrain view renders predicted deformation over mock-positioned nodes, with a time scrubber from now into the model forecast.

Human-verified escalation

Every alert reaches a person through dashboard, email, SMS, then voice, and nothing is permanently auto-dismissed without confirmation.

How it works

1
Sense

ESP32 nodes sample tilt, vibration, displacement

2
Fuse

Gateway combines evidence into one risk score

3
Predict

GNN estimates trend and time-to-threshold

4
Verify

A human confirms: real, blast, or uncertain