WaterLens open-hardware water monitor runs Edge Impulse AI on Arduino UNO Q

WaterLens is a compact, low-cost water quality monitoring device designed for communities without ready access to laboratory testing. The kit uses an Arduino UNO Q microcontroller with a Seeed Studio Grove Base Shield and multiple Grove sensors to measure pH, turbidity, total dissolved solids, and oxidation-reduction potential. A Grove OLED display and dual-button module provide local user interaction without requiring a separate computer or phone. The device runs an Edge Impulse-trained machine learning model locally to flag potential anomalies in sensor readings, though its creator explicitly states it is not a replacement for validated laboratory water quality testing.

The project is designed for assembly without soldering or advanced technical skills, using only off-the-shelf components that connect through the Grove modular ecosystem. This approach lowers barriers to deployment in resource-limited settings where water quality data is needed but professional testing infrastructure is unavailable. The local inference model processes sensor data on-device rather than transmitting to cloud services, which addresses connectivity limitations and privacy concerns in remote monitoring scenarios.

Hardware architecture and edge AI implementation

• Arduino UNO Q core: The device is built around the Arduino UNO Q board, which serves as the main microcontroller for sensor acquisition, local inference, and display management.
• Grove modular sensor ecosystem: pH, turbidity, TDS, and ORP sensors connect via a Seeed Studio Grove Base Shield, enabling solderless assembly and swapability of components.
• Edge Impulse model deployment: The anomaly detection model is trained and deployed through Edge Impulse's platform, running inference locally on the UNO Q without network dependency.
• Local display and input: A Grove OLED screen and dual-button module provide standalone operation, eliminating the need for paired smartphones or computers in the field.
• Explicit non-diagnostic scope: The firmware and design are intentionally scoped to supplemental anomaly flagging rather than safety certification, with clear documentation of this limitation.

Why it matters

WaterLens demonstrates a reproducible pattern for environmental sensing projects in the open-hardware community, combining commodity Arduino hardware with modern edge ML tooling. Its solderless Grove-based construction and explicit documentation of limitations set a useful precedent for responsible open-source hardware deployment in health-adjacent applications. For Linux and embedded developers, the project highlights how Edge Impulse's model compression and deployment pipeline can bring neural network inference to resource-constrained 8-bit and 32-bit microcontrollers without requiring custom RTOS ports or mainline kernel drivers.