https://github.com/jkanishkha0305/autodrive-vision

Multi-model approach for autonomous driving 🚗🤖: A holistic exploration of traffic sign detection 🛑🚦, vehicle detection 🚗📡, and lane detection 🛣️📸, powered by the magic of deep learning 🧙‍♂️, within the captivating world of the Udacity Self-Driving Car Simulator 🚀🎮.
https://github.com/jkanishkha0305/autodrive-vision

Last synced: 4 days ago
JSON representation

Multi-model approach for autonomous driving 🚗🤖: A holistic exploration of traffic sign detection 🛑🚦, vehicle detection 🚗📡, and lane detection 🛣️📸, powered by the magic of deep learning 🧙‍♂️, within the captivating world of the Udacity Self-Driving Car Simulator 🚀🎮.

• Host: GitHub
• URL: https://github.com/jkanishkha0305/autodrive-vision
• Owner: Jkanishkha0305
• License: mit
• Created: 2022-11-10T14:18:31.000Z (about 2 years ago)
• Default Branch: main
• Last Pushed: 2023-10-13T17:45:32.000Z (about 1 year ago)
• Last Synced: 2023-10-16T16:11:44.387Z (about 1 year ago)
• Language: Jupyter Notebook
• Homepage:
• Size: 35.2 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# AutoDrive Vision

## Multi-Model Autonomous Driving

🚗 Building the Future of Autonomous Vehicles with Deep Learning 🤖

Welcome to the Multi-Model Autonomous Driving project! Here, we're on a journey to advance the world of self-driving cars using the power of deep learning and sensor fusion.

## Project Overview

In this repository, you'll find a comprehensive exploration of multiple deep learning models designed to enhance autonomous driving capabilities. We're not just building models; we're paving the way for safer, smarter, and more efficient autonomous vehicles.

### Key Achievements

- **Multi-Model Approach:** 📊 We've developed and tested various deep learning models to detect and classify traffic signals, spot obstacles, and identify lanes.

- **Comparative Study:** 📈 We've conducted in-depth comparative studies using prominent models like Mask-RCNN, ResNet50, InceptionV3, and MobileNet in realistic simulated environments.

- **3D Data Visualization:** 🌐 Our work includes KITTI 3D data visualization, which plays a pivotal role in understanding the vehicle's surroundings.

- **Algorithm Implementation:** 🤖 We've worked with various cutting-edge algorithms, including FCNN, DeepSort, MTAN, SFA 3D, UNetXST, and ViT, to improve vehicle perception.

- **Real Autonomous Vehicle:** 🚀 We've taken our knowledge and applied it to build a tangible autonomous driving vehicle. This real-world system uses Jetson Nano, Arduino, and Ultrasonic Sensors to detect lanes, avoid obstacles, and respond to traffic signals through deep learning and image segmentation.

## Getting Started

Ready to dive into the world of autonomous driving and deep learning? Check out our project's code, data, and documentation:

- [Code](/code) - Explore the deep learning models and code used in the project.

- [Data](/data) - Access the datasets and data preprocessing scripts.

- [Documentation](/docs) - Dive into our project documentation to understand the algorithms, models, and implementation details.

## License

This project is open-source under the [MIT License](/LICENSE). Feel free to use, modify, and contribute to our work.

## Acknowledgments

We'd like to express our gratitude to the open-source community, researchers, and developers who have paved the way for advancements in autonomous driving and deep learning.

Happy Coding and Safe Driving! 🛣️👨‍💻🚗