Embedded / IoT2026
SAVE YOUR GRANDPA
A wearable fall detector with a four-state confirmation machine
- ESP32
- ESP-NOW
- MPU6050
- HC-SR04
- Python
- SQLite
- REST
A chest-worn ESP32 watches motion and impact and reports over ESP-NOW to a base station that logs telemetry and serves a dashboard. The detector only confirms a fall after four conditions line up in sequence, which is what separates a real fall from sitting down hard.
- 01
Four-state fall confirmation
NORMAL → MOTION_EVENT → IMPACT → FALL_CONFIRMED. A fall is only declared when motion exceeds threshold, an impact follows shortly after, chest orientation changes, and the body then goes still.
- 02
Threshold tuning against real movement
Seven acceleration and gyroscope thresholds were tuned against standing, walking, running, bending and staged falls to push down false positives.
- 03
Two dashboards, one telemetry store
A terminal dashboard for bench work and a remote dashboard server that persists to SQLite and serves a browser UI with REST endpoints for status, history and events.
