LaneBot

LaneBot: Finding the Lane to Smarter Driving.

Comment

Inspiration

The inspiration for LaneBot came from the desire to enhance safety and awareness for older cars that lack modern, built-in AI systems. With the hackathon theme "Timeless Moments Await," we aimed to bridge the past, present, and future by bringing cutting-edge AI technology to classic vehicles. We wanted to make driving safer and more intelligent by providing real-time lane detection, object detection, and live recording, all while ensuring compatibility with older cars. Our goal was to unlock new possibilities for drivers everywhere, regardless of their vehicle's age.

What it does

LaneBot is an AI-powered driving assistant designed to transform older cars into safer, smarter vehicles. It features:

  • Real-Time Lane Detection: Accurately identifies lane boundaries and monitors the car's position to help drivers stay on track.
  • Object Detection & Distance Calculations: Continuously detects objects on the road and calculates their distance from the car, providing critical spatial awareness.
  • Live Dashcam Recording: The app uses a camera to record live footage throughout the journey, highlighting important moments with vibrant colors to make them easily identifiable and reviewable.
  • Visual Feedback: Displays alerts and information through a user-friendly interface to enhance driver awareness and safety.

By integrating these features, LaneBot offers an intelligent, real-time driving assistant that makes everyday driving safer and more intuitive.

How we built it

We built LaneBot using a combination of Swift, Python, and Figma: -Frontend Development: Designed using Figma, where we focused on building a clean, modern, and intuitive user interface that effectively displays real-time data to the driver.

  • Backend Development: Implemented in Swift and Python, allowing us to leverage AI models for lane detection and object detection. -Testing & Simulation: We conducted a real-life simulation by driving and using the app to detect lane boundaries, objects, and their distances in various driving conditions. This allowed us to validate our approach and optimize our algorithms for accuracy.

Challenges we ran into

Building LaneBot required integrating multiple technologies seamlessly.

  • Swift Integration: Implementing real-time AI detection systems with Swift while ensuring compatibility with Python models were challenging.
  • UI Design: Creating a functional, visually appealing interface in Figma that effectively displayed real-time data required multiple iterations.
  • Testing & Simulation: Testing LaneBot in real-world conditions and refining algorithms for accuracy involved rigorous debugging.
  • Feature Prioritization: Balancing ambitious features with practical implementation within a limited timeframe was crucial.

We believe we overcame these challenges through adaptability, focus, and continuous refinement.

Accomplishments that we're proud of

We are incredibly proud of our progress and achievements, particularly in areas where we initially felt underqualified.

  • Learning Swift: Despite our lack of experience, we successfully implemented complex functionalities using Swift, overcoming a steep learning curve.
  • Mastering Figma: Our frontend developers, who were completely new to Figma, managed to create a polished and functional user interface through dedicated research and experimentation.
  • Real-Life Simulation Testing: Conducting a real-life test with LaneBot demonstrated the effectiveness of our approach and provided valuable insights for further improvement.

What we learned

  • Swift Fundamentals: We gained hands-on experience with Swift, understandings its syntax, libraries, and how to integrate AI models.
  • UI/UX Design in Figma: We learned how to effectively use Figma to create intuitive and visually appealing user interfaces.
  • AI Integration: We enhanced our knowledge of AI models for object detection and lane detection, learning how to apply them in real-time environments.
  • Collaboration & Adaptability: Working on a project with unfamiliar tools taught us how to quickly learn, adapt, and collaborate under pressure.

What's next for LaneBot

As a team, we have a few ideas for LaneBot and it includes:

  • Enhanced Object Detection: Implement more advanced models to improve detection accuracy and reduce false positives.
  • Improved Real-Time Feedback: Incorporate voice alerts and haptic feedback for a richer, more interactive experience.
  • Crowdsourced Road Data: Enable users to share road obstacle data to enhance map accuracy and safety warnings. These are some ideas that we had and wanted to implement for the hackathon, but as we changed our expectations to be more realistic. However, that means that we can work on it after the hackathon as we are truly happy about this project.

Built With

Share this project:

Updates