Arduino UNO Q contest winners showcase edge AI across water quality, robotics, spectroscopy

Arduino, Qualcomm Technologies, and Edge Impulse partnered with Hackster to run the Invent the Future competition, which drew 4,028 registrants and yielded 153 final project submissions built around the Arduino UNO Q board. The winners span six categories plus a Best in Show award, with projects ranging from a real-time water quality monitoring system to a wearable roller coaster telemetry rig, a haptic sensory extension headset, and an edge-AI spectrophotometer for material analysis.

The UNO Q serves as the common hardware platform across all winning entries, paired with the Arduino App Lab software environment. Several projects integrate Edge Impulse machine learning models for on-device inference, while others incorporate additional hardware such as SparkFun IMUs and spectroscopy sensors. The competition results demonstrate how the board's dual-core architecture and AI acceleration are being applied to practical problems in environmental sensing, agriculture, accessibility, and scientific instrumentation.

UNO Q hardware and winning project architectures

• Dual-core edge AI platform: The UNO Q is described as a "dual-brain board" that runs Edge Impulse machine learning models locally, enabling inference without cloud dependency in projects like the Smart Water Guardian AI and Edge-AI Spectrophotometer.
• Sensor integration: Winning projects attach diverse analog and digital sensors to the UNO Q, including water quality probes (temperature, pH, turbidity, dissolved solids), a SparkFun 9DoF IMU for motion capture, and a SparkFun AS7343 Triad Spectroscopy Sensor for 11-channel light wavelength analysis.
• Haptic and display I/O: The HaloSense V1 drives 13 haptic feedback motors directly from the UNO Q, while the Curie robot uses an OLED screen for facial expression output and connects to cloud APIs including Groq for LLM-based task assistance.
• Local ML inference patterns: The RCT IRL project runs three separate Edge Impulse models post-collection to generate scalar scores (Excitement, Intensity, Nausea) from IMU data, demonstrating a multi-model inference pipeline on the board.

Why it matters

These projects illustrate how Arduino's Qualcomm-powered UNO Q is positioning itself against Raspberry Pi and NVIDIA Jetson in the low-cost edge AI market, particularly for developers who need on-device inference without Linux system administration overhead. The competition's focus on commercially available components and documented sensor pairings provides a reference library for embedded engineers building environmental monitors, agricultural automation, and accessible assistive devices.