Advanced Driver Assistance System (ADAS) - Jetson Nano Implementation

An advanced driving assistance system based on NVIDIA Jetson Nano embedded computer with comprehensive capabilities for lane detection, traffic sign recognition, and object detection on roadways.

Features

• Lane Detection: Real-time lane boundary detection and tracking
• Traffic Sign Recognition: Detection and classification of traffic signs
• Object Detection: Vehicle and pedestrian detection
• Lane Departure Warning: Vehicle position monitoring relative to lane markings
• Performance Optimized: TensorRT acceleration for real-time processing

Development Environment and Build Requirements

System Requirements

• CMake >= 3.10
• OpenCV with CUDA >= 4.0.1
• C++ 17 Compiler
• CUDA 10.1
• TensorRT 5
• NVIDIA Jetson Nano (recommended)

Setup Instructions for Linux - Ubuntu 18.04/20.04

1. Download Required Data

Pre-trained Models

Download the pre-trained neural network models from Google Drive and place them in the models/ folder.

Test Data

Download test images and videos from Google Drive and place them in the root folder of this project.

2. GPU Architecture Configuration

Update the GPU_ARCHS parameter in CMakeLists.txt to match your GPU architecture:

• Jetson Nano: GPU_ARCHS = 53
• Other GPUs: Check your GPU's compute capability at NVIDIA GPU Compute Capabilities

For detailed information on GPU architectures:

• How to get NVIDIA graphics card GPU_ARCHS for Linux
• Matching CUDA arch and CUDA gencode for various NVIDIA architectures

3. Compilation and Execution

Build the Project

cd <project_directory>
mkdir build
cd build
cmake ../
make

Run the Application

./ADAS

Installation Help

If you encounter issues with the standard installation procedures, refer to these helpful resources:

CUDA 10.1 Installation

Installing CUDA 10.1 on Ubuntu 20.04

OpenCV with CUDA Installation

How to install OpenCV 4.2.0 with CUDA 10.0 in Ubuntu distro 18.04

Project Structure

ADAS/
├── main.cpp                 # Main application entry point
├── CMakeLists.txt          # Build configuration
├── README.md               # This file
├── LICENSE                 # Apache 2.0 license
├── configs/                # Configuration files
├── common/                 # Common utilities and models
├── lane_detection/         # Lane detection implementation
├── lane_detection_NN/      # Neural network training scripts
├── models/                 # Pre-trained model files
└── input_folder/           # Test images and videos

License

This project is licensed under the Apache License 2.0. See the LICENSE file for details.