Inspiration
In many remote areas and under-resourced countries, limited access to emergency services and immediate care often leaves accidents unnoticed, putting lives at risk. Our goal is to bridge this gap by bringing safety directly to the driver’s phone. Using AI, our system is designed to detect and assess accidents in real-time, ensuring that any harm or sudden danger to the driver is quickly recognized, allowing for a rapid response when it’s needed most. This solution empowers drivers with enhanced safety, regardless of where they are.
What it does
Our app, named G.U.A.R.D. to reflect its role as a vigilant guardian, leverages current weather conditions and user driving habits to assess risks and provide simple, proactive safety recommendations tailored to situations like rain or snow. While this feature enhances the overall driving experience, the app's primary function is to ensure safety during accidents. It uses real-time location tracking to monitor speed and acceleration, recognizing patterns indicative of a potential crash. When the app detects an unusual or sudden halt not explained by rapid acceleration, it activates an AI chatbot to engage with the driver. If the driver doesn’t respond within a designated timeframe, the app automatically alerts first responders, sharing the driver’s precise location to expedite assistance and retrieval.
How we built it
We used a combination of Flutter for mobile development and Fetch.ai to create a robust and responsive system for real-time emergency response. Google Gemini provided the natural language processing capabilities necessary to add an emotional component to the chatbot interactions, ensuring empathetic communication during stressful events. The app communicates with multiple agents to monitor critical metrics like speed, location, and heartbeat, seamlessly integrating Google Cloud's Speech-to-Text API to transcribe the driver’s voice and assess their well-being. By connecting these technologies, G.U.A.R.D. delivers a proactive safety solution that’s both adaptive and reliable.
Challenges we ran into
We encountered several challenges throughout the development process. The most significant hurdle was managing real-time data processing and accurate accident detection without compromising the app’s speed and responsiveness. Additionally, integrating multiple AI components and ensuring seamless communication between the Fetch.ai and Google Gemini agents required meticulous troubleshooting. Another challenge was achieving consistent performance across varied network conditions, as our target audience includes drivers in remote or under-served areas with limited internet connectivity.
Accomplishments that we're proud of
One of our proudest accomplishments is creating a solution that has the potential to save lives, particularly in areas with limited emergency resources. We’re thrilled with the accuracy and responsiveness of our crash detection system, which is capable of identifying even subtle signs of an accident. Additionally, we’re proud of the seamless integration between Fetch.ai and Google Gemini, enabling our app to interact with users in a supportive and empathetic manner. This project also pushed us to master new technologies and improve our problem-solving skills, resulting in a finished product that we are proud to present.
What we learned
Developing G.U.A.R.D. taught us the importance of balancing multiple technologies to create a cohesive user experience. We gained hands-on experience with real-time data processing, and improved our understanding of how to design apps that can adapt to unpredictable situations. We also learned a great deal about AI-driven communication, honing our ability to build an empathetic user interface. The project provided insights into handling real-world challenges, such as low connectivity and variable GPS accuracy, which are crucial factors for a reliable safety app.
What's next for G.U.A.R.D.
Moving forward, we aim to expand G.U.A.R.D.’s functionality by incorporating additional safety features, such as weather-based alerts and predictive analytics to warn users of potential hazards. We’re also planning to integrate health monitoring capabilities using wearable devices to capture vitals like heart rate and stress levels. Our goal is to improve the app’s ability to assist drivers in remote areas by exploring offline capabilities and increasing its adaptability across different mobile networks. Additionally, we hope to collaborate with emergency services to ensure that G.U.A.R.D. can directly notify responders in under-resourced regions, further strengthening its life-saving potential.
Log in or sign up for Devpost to join the conversation.