This repository documents the multi-node hardware topology — mesh VPN networking (e.g. Tailscale), a host control-plane workstation, a cloud GPU simulation node, and a Jetson physical edge node — used to run autonomous sim-to-real robotics pipelines.
This is an architecture and reference writeup, not a runnable system. It captures reusable, public-safe setup patterns for a distributed sim-to-real pipeline: an Ubuntu / ROS 2 host workstation, a GPU-accelerated cloud simulation node (GCP), and an NVIDIA Jetson Orin edge device. There is nothing to install or execute from this repo; each document describes the pattern and the verification steps for one node or layer.
For repo-specific working rules, read docs/OPERATING_STANDARD.md. For definitions of key terms, see the AI & Robotics Glossary.
| Tier | Node | Role |
|---|---|---|
| Host | Ubuntu / ROS 2 workstation | Primary development node: source control, build tooling, local sim. |
| Cloud | GPU VM on GCP | Burst GPU capacity for physics simulation and synthetic data. |
| Edge | NVIDIA Jetson Orin | Runs robotics workloads close to sensors, actuators, and rigs. |
| Mesh | Private network / VPN overlay | Connects all nodes so ROS 2 discovery and topics work across tiers. |
Each document is a self-contained setup pattern with a verification section:
QUICKSTART.md: How the pieces fit together and where to start.docs/HOST_WORKSTATION.md: Host workstation baseline and setup pattern.docs/CLOUD_SIMULATION_NODE.md: Cloud GPU simulation node provisioning pattern.docs/JETSON_EDGE_NODE.md: Jetson Orin edge node setup pattern.docs/ROS2_NETWORKING.md: ROS 2 discovery, DDS, and cross-network considerations.docs/OPERATING_STANDARD.md: Working rules and quality bar for this repo.
| Component | Version / Type | Purpose |
|---|---|---|
| Operating System | Ubuntu 22.04 LTS | Standard for robotics development |
| Robotics Middleware | ROS 2 Humble | Core communication and tooling framework |
| GPU Driver | NVIDIA Proprietary Driver | Enables GPU acceleration for AI / simulation |
| Simulation Platform | Isaac Sim | High-fidelity physics simulation & sensor data |
| Edge AI SDK | JetPack SDK | OS & libraries for the Jetson platform |
| Version Control | Git | Tracking changes and managing project history |
| Code Hosting | GitHub / gh CLI |
Publicly showcasing and managing repositories |
| Build Tool | Colcon | Building ROS 2 packages and workspaces |
The documented setup workflow reflects core competencies in:
- Systems Administration: Installing and configuring a Linux (Ubuntu 22.04) environment from scratch, including disk partitioning for dual-boot systems.
- Hardware & Driver Management: Installing and verifying proprietary NVIDIA drivers on Linux, including dependency handling and Secure Boot (MOK) enrollment.
- Distributed Systems & Networking: Establishing and verifying a multi-machine ROS 2 network, demonstrating an understanding of the DDS discovery mechanism.
- Embedded & Edge AI Systems: Flashing and configuring an embedded device (NVIDIA Jetson) with the JetPack SDK for sim-to-real deployment.
- Version Control & Technical Documentation: Using Git and GitHub to maintain structured documentation for a technical project.
This project is licensed under the Apache 2.0 License. See the LICENSE file for details.