UWFSAE Jetson ROS2 Stack

Driverless Formula Student autonomous stack running ROS 2 Jazzy on NVIDIA Jetson Orin AGX.

Overview

This repository provides a Docker-based development workflow for building and testing driverless vehicle software. The same codebase runs on both local development machines and the NVIDIA Jetson Orin AGX deployment target.

Key Features:

• Containerized development with Docker 🐳
• ROS 2 Jazzy with pre-configured environment
• Identical environments across development and deployment

Getting Started

New to the project? → See docs/SETUP.md for complete setup instructions.

Quick reference:

# Build container
cd docker && docker build -t jetson-ros2:latest -f Dockerfile ..

# Start container (development machines)
docker-compose up -d

# Enter container
docker-compose exec driverless bash

# Build workspace (inside container)
cd /workspace/ros2_ws && colcon build && source install/setup.bash

# For Jetson with GPU:
# docker-compose -f docker-compose.yml -f docker-compose.jetson.yml up -d

(Temporary) Project Structure

jetson-ros2/
├── docker/                    # Docker configuration
│   ├── Dockerfile             # ROS 2 Jazzy container
│   └── docker-compose.yml     # Compose configuration
├── ros2_ws/                   # ROS 2 workspace
│   ├── src/
│   │   └── demo_pkg/          # Demo package
│   │       ├── demo_pkg/      # Python package
│   │       │   ├── demo_orchestrator_node.py
│   │       │   ├── demo_perception_node.py
│   │       │   ├── demo_planning_node.py
│   │       │   └── demo_control_node.py
│   │       ├── package.xml    # ROS 2 package manifest
│   │       ├── setup.py       # Python package setup
│   │       └── test/          # Unit tests
│   ├── build/                 # Build artifacts (gitignored)
│   ├── install/               # Install space (gitignored)
│   └── log/                   # Build logs (gitignored)
├── docs/                      # Documentation
│   └── SETUP.md               # Setup instructions
├── scripts/                   # Utility scripts
└── README.md                  # This file

Demo Architecture

The autonomous system follows a standard robotics pipeline:

┌─────────────────┐
│  Orchestrator   │  ← Coordinates system state
└────────┬────────┘
         │
    ┌────┴────┬─────────┬─────────┐
    │         │         │         │
┌───▼────┐ ┌─▼──────┐ ┌▼──────┐ ┌▼────────┐
│Perception│ │Planning│ │Control│ │ (More) │
└──────────┘ └────────┘ └───────┘ └─────────┘

Current Demo Nodes:

• demo_orchestrator_node - System coordinator and state machine
• demo_perception_node - Sensor data processing (simulated)
• demo_planning_node - Path planning and decision making
• demo_control_node - Vehicle control commands

All nodes communicate via ROS 2 topics and follow the Jazzy API.

Development Workflow

1. Edit code on your local machine in ros2_ws/src/
2. Rebuild inside container: colcon build --packages-select <pkg>
3. Test immediately - changes are live via volume mounting

Docker Environment

The same Dockerfile builds identical environments for development and Jetson deployment - same OS, ROS 2, Python, and dependencies. On Jetson, we add GPU runtime and hardware access via an override file.

Development (Mac/Windows/Linux):

cd docker
docker-compose up -d
docker-compose exec driverless bash

Jetson (adds GPU/hardware):

cd docker
docker-compose -f docker-compose.yml -f docker-compose.jetson.yml up -d
docker-compose exec driverless bash

Use docker-compose to manage containers - it allows multiple terminals to enter the same container, essential for running multiple nodes.

See docs/SETUP.md for complete instructions.