Vayutech

Yes

Yes

GRU forecasting GRU Auto encoder

PMS7003,BME680/DHT22,MH-Z19,MQ131,MQ135,MQ2,Mics-6814,

None

None

Air pollution is a major problem because conventional monitoring systems often fail to identify local pollution sources or predict worsening air quality. VayuTech addresses this challenge through real-time, multi-node air-quality monitoring, Edge AI-based pollution source identification, anomaly detection, and forecasting using Arduino UNO Q. It aims to provide timely alerts and support automatic air filtration.

PyTorch on Python, running locally on the Linux side of Arduino UNO Q. VayuTech uses three Gated Recurrent Unit (GRU) models: a pollution forecasting model, a pollution source classification model, and an anomaly detection autoencoder. The models are designed for local training and inference using collected sensor data without requiring cloud connectivity.

Main processor (MPU) -Qualcomm Dragonwing QRB2210 Microcontroller (MCU) -STM32U585 RAM - 4 GB LPDDR4 Internal storage - 32 GB eMMC Wireless connectivity -Wi-Fi and Bluetooth Low Energy (BLE) Camera interfaces -USB camera and 2 × MIPI CSI camera Display interface- Modulino Input peripherals -Keyboard and mouse

The system is designed for deployment on Arduino UNO Q using Edge AI for local air-quality analysis and real-time decision-making. Multiple sensor nodes communicate wirelessly using LoRa, while the central unit processes pollution data, predicts air-quality trends, displays results on a dashboard, and activates the filtration system when required. The prototype currently includes a software-based simulation, with hardware integration planned for practical deployment.

The key innovation of VayuTech is the integration of Edge AI, multi-node air-quality monitoring, pollution forecasting, source identification, and automatic air purification in one system. It aims to provide faster local decisions, reduce dependence on cloud connectivity, and support early pollution alerts with a modular three-stage filtration system consisting of pre-filtration, HEPA, and activated carbon filters.

VayuTech is an Edge AI-based air-pollution monitoring and filtration solution designed to detect, analyze, and respond to changing air quality. Multiple sensor nodes collect pollution data and transmit it through LoRa to a central processing unit based on Arduino UNO Q. AI models analyze pollution levels, forecast trends, and help identify possible pollution sources. The system displays air-quality status and recommendations on a dashboard and is designed to activate a three-stage filtration unit when pollution levels are high. Its goal is to enable intelligent, timely, and locally processed air-pollution management.