SNOM is an emotionally intelligent, AI-powered robotic companion designed to support neurodiverse children—especially those with autism—in developing social skills, emotional understanding, and self-confidence.

Built to act as a best friend, SNOM follows children around, engages in adaptive conversations, and provides a safe, non-judgmental presence throughout their day.

Developed at Bitcamp 2025, SNOM blends cutting-edge tech with deep empathy to empower children and give parents peace of mind.

The silent struggles of neurodiverse children navigating a world that often feels overwhelming—and the quiet heartbreak of parents who long to help—led us to create SNOM.

We envisioned a gentle buddy who always listens, never judges, and understands kids just the way they are.

SNOM is born from empathy, built with intention, and powered by the belief that every child deserves a friend who truly sees them.

SNOM is a smart, mobile, and empathetic robot that:

• 🤝 Autonomously follows children using real-time computer vision for companionship and safety.
• 📏 Maintains optimal conversation distance using dynamic calculations.
• 🧠 Uses Gemini 2.0 Flash to engage in natural, mood-aware conversations.
• 🎭 Visually displays emotions on a screen to encourage social understanding.
• 🔊 Speaks using lifelike voice feedback powered by text-to-speech tech.
• 🧭 Navigates independently to locate and follow users while avoiding obstacles.
• 🌱 Helps children identify, express, and manage emotions through conversation and interaction.

• Raspberry Pi 5 – Brain of SNOM, coordinating vision, movement, and speech.
• Modified RC Car Chassis – Provides smooth and flexible mobility.
• Camera Module – Enables person detection and tracking using OpenCV.
• Bluetooth Speaker – Provides natural voice feedback.
• Display Screen – Shows SNOM’s emotional expressions.
• Power Bank (10,000mAh) – Ensures extended runtime on the go.

• Vision System – Built with OpenCV for object tracking and real-time distance estimation.
• Motor Control – Uses RPI.GPIO for movement commands.
• Conversation AI – Integrated with Gemini 2.0 Flash for dynamic dialogue.
• Speech Output – Handled via pyttsx3 for engaging voice interaction.
• Emotion Display – HTML/CSS/JS frontend served via Flask to display expressive emotions.

• Fine-tuned GenAI models to support emotional awareness in conversations.
• Built personalized child personas for meaningful, relatable interactions.
• Helped us understand real-world needs of parents and children with autism.
• Supported the development of an empathy-first solution.
• Enhanced our business plan and future scalability roadmap.

• 🔩 Hardware Firsts – Tackling hardware as CS students was a bold step!
• 🔧 Reverse Engineering – Modifying an RC car without compromising its internals.
• ⚡ Power Optimization – Managing multiple modules running concurrently.
• 🧠 Real-Time Logic – Ensuring smooth decision-making and movement.
• 🔈 Audio Clarity – Maintaining speech quality in different environments.

• 🛠️ Fully reverse-engineered and repurposed an RC car for autonomous tracking.
• 🤖 Integrated AI, speech, vision, and emotional expression into one cohesive system.
• 🧠 Developed mood-based conversations powered by Gemini 2.0 Flash.
• 💬 Built an engaging, empathetic tool that truly connects with children.
• 👪 Created a product that gives peace of mind to parents while fostering growth in kids.

• How to merge software logic with real-world hardware mechanics.
• Practical experience with real-time vision processing under constraints.
• Human-robot interaction design and emotion-centered UX.
• Power management and optimization techniques.
• Most importantly: empathy-first design transforms lives.

• 🌍 Global Reach – Make SNOM accessible to families worldwide.
• 🧠 Advanced Personalization – Adapt SNOM to unique personalities.
• 👩‍⚕️ Therapeutic Integration – Partner with educators and therapists.
• 📱 Mobile App – Let parents monitor and customize SNOM via a companion app.
• 🔋 Better Battery Life – Optimize SNOM’s energy usage.
• 🧠 Smarter Conversations – Continuously evolve emotional intelligence models.

• Raspberry Pi 5 (with Raspberry Pi OS)
• Python 3.9+
• OpenCV 4.5+
• Chromium browser (for emotion display)
• RC car chassis (modified)
• Compatible Pi camera module
• Bluetooth speaker
• Display screen (7"+ recommended)
• Power bank (10,000mAh or higher)