Inspiration
With increasing urban pollution and traffic congestion, we were motivated to create a system that encourages Eco-friendly driving and reduces harmful emissions. The idea came from observing how daily driving behaviours directly impact air quality and sustainability. We wanted to use Edge AI to make real-time, data-driven decisions that promote cleaner, smarter urban transport.
What it does
Our system uses Edge AI devices to collect vehicle and environmental data — such as speed, acceleration, and emission levels — in real time. It analyses the data locally to: Detect inefficient or harmful driving behaviours. Provide real-time feedback to drivers for Eco-driving improvement. Map pollution levels across urban regions. Suggest optimised routes that minimise fuel consumption and emissions.
How we built it
We integrated multiple technologies to bring this project to life: Edge AI Hardware: Used Raspberry Pi/NVIDIA Jetson for on-device processing. Sensors: Gathered vehicle performance and air quality metrics (CO₂, NO₂, PM2.5). Machine Learning Models: Trained models to predict emission impact and driving efficiency. Cloud Dashboard: Aggregates data for city-wide pollution visualization. Mobile Interface: Provides real-time driver insights and eco-scores.
Challenges we ran into
Processing complex data efficiently on low-power edge devices. Synchronizing edge and cloud data without latency. Collecting reliable sensor data in varying weather and traffic conditions. Designing models that balance performance and accuracy for real-time feedback.
Accomplishments that we're proud of
Successfully implemented real-time pollution intelligence using Edge AI. Built a working prototype that provides driver feedback within seconds. Achieved noticeable reductions in simulated CO₂ emissions. Created an interactive dashboard visualizing live pollution and route efficiency.
What we learned
How to deploy AI models on edge devices for real-time decision-making. The importance of data preprocessing and optimization in IoT systems. How driver behavior analysis can significantly influence sustainability. Collaborative problem-solving and system integration skills.
What's next for Edge AI-Powered Eco Driving and Pollution Intelligence
We plan to: Integrate the system with city-wide smart transportation grids. Improve ML model accuracy using deep learning and larger datasets. Partner with automotive companies and municipalities to deploy it in real-world conditions. Add features like predictive maintenance and AI-powered traffic management for full smart mobility support.
Built With
- amazon-web-services
- and
- cloud
- edge-ai-hardware-(raspberry-pi-/-nvidia-jetson-nano)
- iot
- iot-sensors
- like
- node.js
- python
- react.js
- services
- tensorflow
