Guard every hive.
Save the harvest.
BeeGuardian AI is a low-cost embedded platform that fuses acoustic TinyML with temperature, humidity and weight sensors to catch colony problems days before a beekeeper could.
- 148ms
- Edge inference
- 0–100
- Hive Health Score
- 24/7
- Solar-powered
The platform
Four senses. One score.
Sensor fusion turns raw signals into a single Hive Health Score so beekeepers act on insight, not noise.
Acoustic Edge AI
MFCC + neural network on the ESP32 classifies Normal, Stress and Noise from bee sounds in under 150 ms.
Sensor fusion
Temperature, humidity, weight and audio confidence combine into a single 0–100 Health Score.
MQTT real-time
Telemetry streams to HiveMQ. The dashboard updates live. Remote commands flow back to the hive.
Live analytics
Beautiful, animated charts for weight trends, foraging gains, thermal stability and colony sentiment.
Solar autonomy
18650 + TP4056 + solar panel. Deep sleep between reads. Weeks of runtime without sun.
Early alerts
Popup, timeline and email-ready alerts for heat, humidity spikes, rapid weight loss and stress buzzing.
Food security
Healthy pollinators mean stable harvests. BeeGuardian protects colonies that protect our food.
Open & modular
Arduino firmware, Edge Impulse project, MQTT topics and web dashboard — all documented and hackable.
Hardware
Rugged, solar, under $40 in parts.
Off-the-shelf components, weatherproof enclosure, deep-sleep friendly firmware. Designed for beekeepers in the field — not lab benches.
- •ESP32 DevKit V1
- •INMP441 I²S microphone
- •BME280 environmental sensor
- •Load cell + HX711 amplifier
- •5 W solar panel
- •18650 Li-Ion + TP4056 charger
- •Optional RTC DS3231
- •IP65 weatherproof enclosure
