HeliosAI

Admin dashboard for user to get info from the sensors on arduino.
When talking to the agent, an audio visualizer will appear.
Learn more about what Helios's mission.
Chat with snowflake ai to get information on solar energy and our mission

Inspiration

We are battling the rise in electricity costs by developing a more eco-friendly and sustainable energy source. Our goal is to make solar systems smarter and more efficient through automation and AI.

What it does

HeliosAI uses a solar cell connected to an Arduino with light sensors and servo motors. It can rotate 180° on both the X and Y axes, allowing it to track the brightest light source automatically. The data collected—such as light intensity and energy generated—is sent to a Flask web dashboard powered by an AI agent that displays real-time performance. The system also includes a conversational AI agent that communicates this data through voice interaction.

How we built it

We used:

Arduino for sensor input and motor control.
Flask for the web dashboard and API endpoints.
Google ADK (Gemini) and Cloud API for the AI agents that fetch and relay data.
Serial communication between Arduino and Flask to send sensor data continuously.

The AI agent on the dashboard processes this data and shares it with another conversational agent, enabling audio feedback for users.

Challenges we ran into

Integrating agent-to-agent communication using the Google ADK Gemini SDK.
Calibrating the Arduino light sensors and ensuring smooth dual-axis movement.
Managing real-time data flow between hardware and the web dashboard.

Accomplishments that we're proud of

Built a functioning solar tracker that physically rotates toward the brightest light.
Established a two-way communication system between the AI dashboard and a voice agent.
Combined hardware, AI, and software integration in one project within hackathon time limits.

What we learned

How to program and wire an Arduino from scratch.
How to connect hardware with a web app using Flask and APIs.
How to integrate multiple AI agents for data processing and interaction.

What's next for HeliosAI

We plan to:

Upgrade to a more robust Arduino model with higher torque servos.
Improve sensor precision and introduce predictive tracking using machine learning.
Expand the dashboard with live analytics and cloud