Project summary: 

This project proposal aims at developing the platform with an affordable device to monitor air and noise pollution in India at the personal/community level. The platform integrates pollution sensors into a single board and uses a TinyML microcontroller for on-device  processing. This device connects to the user's mobile phone via Bluetooth Low Energy and provides alerts and personalized recommendations based on the user's health profile and pollution exposure levels. The UI will use structured questionnaires to keep the steps logical and explainable. This device is a significant step in addressing the feature requirement for an environment pollution monitoring device in India at the personal/community level. With the advances in Large Language Models (LLMs) and Natural Language Understanding (NLU), especially in low-resource languages, the idea is to integrate such features using the current state of the art.

Expected Deliverables:

1. Integrating an affordable air and noise pollution monitoring device

2. Developing reliable sensing using AI models to ensure sensor fault tolerance, sensor calibration and activity recognition

3. Implementing intelligence at the Edge

4. Interactive pollution exposure estimation and action recommender

PI & Co-PIs

Collaborators