As engineering students, we constantly face high-stress situations, exams and deadlines that seem impossible in the moment. When your heart is racing and your thoughts are scattered, all you want is someone calm beside you saying, “It’s okay. Do this next.” We realized survival situations are no different, stress clouds judgment, and small mistakes can become dangerous. So we built Twin, a built-in friend designed to guide you when things get overwhelming.
What it does
Twin is an AI-powered survival assistant integrated directly into a hard hat.
It sees your surroundings through a camera, monitors biometric data like heart rate, and provides real-time guidance whenever you need it.
The interaction works like this:
- The user presses a button and asks a question
- Twin captures the surrounding environment using a camera
- Processes the visual input
- Analyzes the user’s question
- Responds with context-aware survival guidance
How we built it
Twin runs on a hybrid embedded system combining:
- Arduino for sensor input (heart rate + environmental data)
- Raspberry Pi for processing and AI inference
- Computer vision software for environmental recognition
- Serial communication between Arduino and Pi for real-time data exchange
When the button is pressed, the Arduino signals the Raspberry Pi. The Pi captures an image, processes it using AI software, interprets the user’s question, and generates a response.
Challenges we ran into
- Our biggest challenge was establishing reliable serial communication between the Arduino and Raspberry Pi. Baud rate mismatches caused inconsistent data transmission, resulting in delayed or corrupted readings.
- We also had to balance processing speed with wearable hardware limitations.
Accomplishments that we're proud of
- Successfully creating a hardware-triggered AI interaction system
- Achieving real-time environmental analysis
- Integrating embedded systems with AI processing
- Building a functional wearable prototype within limited time
What we learnt
- How to configure and program a Raspberry Pi for AI processing
- How to interface Arduino hardware with Linux-based systems
- How critical baud rate matching and serial debugging are in embedded systems
- The challenges of integrating hardware and AI in real time
- Most importantly, we learned how powerful human-centered AI can be when it augments decision-making instead of replacing it.
What's next for Twin
- Continuous multi-frame contextual awareness (always-on perception instead of push-to-talk)
- Improved hazard detection models
- Optimized latency and power efficiency
- Expanded applications in construction safety and disaster response
Twin started as the friend we wished we had during exams. Now, it’s a system designed to stand beside you when it matters most.
Log in or sign up for Devpost to join the conversation.