This repository contains the complete software stack for our autonomous vehicle platform, organized into specialized submodules.
Our autonomous vehicle platform integrates multiple computing systems, sensors, and machine learning models to provide a comprehensive solution for autonomous driving research and development. The system architecture includes:
- Jetson Nano for main vehicle control and autonomous capabilities
- Raspberry Pi for instrumentation cluster and user interface
- Arduino microcontrollers for sensor input and actuator control
- Machine learning models for perception tasks
- Dataset collection and management tools
to check all the features in a visual way.
Team02-Course/
├── JetsonNano/ # Main vehicle control software
├── RaspberryPi/ # Instrument cluster and UI
├── MicroController/ # Arduino code for sensors/actuators
├── MachineLearning/ # ML models for perception
├── Dataset/ # Dataset management and generation
├── ADAS_SIL/ # A Software-in-the-Loop (SIL) testing environment
└── Libs/ # Shared libraries and utilities for cross-module functionality
Contains all the code running on the Jetson Nano, which serves as the main computing platform for the vehicle. Includes:
- Vehicle control systems
- Sensor integration
- Communication frameworks
- Autonomous driving capabilities
- Hardware abstraction layers
Contains code for the Raspberry Pi, which manages the instrument cluster and user interfaces:
- Dashboard displays
Arduino-based software for hardware-level interactions:
- Speed sensor data collection
- Lighting control systems
- Real-time sensor processing
Machine learning models and algorithms for autonomous perception:
- Lane detection models
- Object detection and classification
Tools and resources for dataset management:
- Local track dataset collection tools
- Carla simulator dataset utilities
- Dataset preprocessing pipelines
- Future: Code for generating synthetic datasets using Carla
This project provides a comprehensive testing framework for validating lane detection and object detection algorithms in the CARLA simulation environment:
- Integration with CARLA simulator for realistic driving scenarios
- ONNX-based machine learning models for lane and object detection
- Real-time visualization with Pygame
- Zenoh-based communication for distributed processing
Each submodule contains its own documentation with specific setup instructions. To clone the repository with all submodules:
git clone --recursive https://github.com/SEAME/Team02-Course.git