AI sensor payloads

AI helps map sensor values from raw payloads.

IoT sensors often send JSON payloads with different names for the same measurements. GrowGuard can inspect an example payload and propose which field means temperature, humidity, battery, EC, pH, soil moisture, or another metric. The user validates the mapping before saving, and the system does not invent readings.

TTN, MQTT, NB-IoT User-validated mapping No invented readings
GrowGuard sensor panel with interpreted payload values

Different payloads, one dashboard

Whether data comes from TTN, MQTT, or NB-IoT, GrowGuard works to bring readings into the same operational structure.

Transparent AI suggestions

AI proposes clear metric keys and can expose confidence when a payload is ambiguous.

Human validation

A mapping becomes active only after the user confirms it, preventing a wrong field from automatically entering the dashboard.

Compatible with existing decoders

Stable decoders and flows remain the foundation. AI is especially useful for onboarding and new or unusual JSON payloads.

From raw payload to useful readings

01

Send an example

The sensor sends a JSON payload through TTN, MQTT, NB-IoT, or a compatible flow.

02

AI proposes mapping

GrowGuard identifies possible fields for visible metrics: temperature, humidity, battery, EC, pH, or soil readings.

03

Save what you validate

The user confirms the good mapping and ignores metadata, id, timestamp, or counter fields.

Onboard new sensors without losing days in payload formats.

Use GrowGuard for IoT onboarding and controlled mapping of sensor readings.

Frequently asked questions

Can AI change existing sensor values?

No. AI only proposes how fields should be read; values come from the payload, and existing decoders remain prioritized.

What happens to ids, timestamps, or counters?

Metadata fields are ignored when they do not represent a user-visible measurement.