Skip to content

ShadowHound Knowledge Base

Purpose

Central entry point for the ShadowHound documentation vault. This autonomous mobile robot system combines ROS2 navigation with LLM/VLM-driven task planning on the Unitree Go2 platform.

See Architecture Documentation for system diagrams and design overview.

Prerequisites

  • Clone the repository and open the /docs directory as an Obsidian vault.
  • Install the Obsidian Markdown Links core plugin (enabled by default).

Steps

  1. Review the directory hubs below to explore specific topic areas.
  2. Use the reusable template at _templates/page when authoring new content.
  3. Store shared media inside _assets/ and reference them with relative paths.

Documentation Directories

📋 Planning & Status

Project Overview - Strategic planning, roadmaps, and status tracking - Quick start guides and operational references - 8 active documents: roadmap, todo, setup status, architecture review

🏗️ System Design

Architecture - System architecture and design decisions - Component interactions and data flows - Layered design: Application → Agent → Skills → Robot

🔧 Development

Development - Contributor guides and development policies - Git workflows and submodule management - 11 documents including cleanup tracking and checklists

� Deployment

Deployment - Environment strategies and configuration - Launch orchestration and machine setup - Multi-machine synchronization procedures

�💻 Software

Software - ROS2 packages and implementations - LLM integration (vLLM, Ollama) - 26+ docs - Web interface and WebRTC communication - Configuration and environment setup

🤖 Hardware

Hardware - Unitree Go2 platform and sensor suite - Network and power topologies - Hardware specifications and integration guides

🌐 Networking

Networking - ROS2 DDS and WebRTC configuration - Direct connectivity testing - Multi-machine deployment topologies

🐛 Troubleshooting

Troubleshooting - Diagnostic workflows and health checks - Known issues and resolutions - Startup validation and robot testing

🔬 Research & Testing

Simulation - Gazebo and hardware-in-the-loop Research - Experiments and development logs Performance - Benchmarking and optimization

📦 Supporting

Integrations - DIMOS, vision, AI integrations Known Issues - Bug tracking and workarounds History & Archive - Archived legacy documentation

Validation

  • [ ] Vault opens in Obsidian without warnings.
  • [ ] All category links resolve inside Obsidian.
  • [ ] Link conversion step renders correctly on GitHub Pages and Wiki.

References