Welcome to Bob’s Handbuch

This is a collection of various ROS 2 packages and nodes for natural language processing, LLM integration, and system control. They leverage ROS topics to connect NLP and LLM components seamlessly.

Key Features:

  • LLM Integration: OpenAI-compatible API interface with stateful conversations and dynamic tool calling.

  • Speech & Audio: High-fidelity TTS (Qwen3, XTTS), offline STT, and multi-source audio mixing/conversion.

  • Vision & VLM: Visual reasoning, captioning, and object detection via Moondream2 VLM.

  • Image Generation: Modern Text-to-Image and Image-to-Image generation (Stable Diffusion 3.5, FLUX.2).

  • Visualization: High-performance SDL2 rendering and headless streaming (Twitch) with dynamic overlays.

  • Vector Database: Semantic storage and multimodal retrieval using Chroma and Qdrant.

  • Human-Machine Interface: Sentiment-driven facial animations and interactive web-based GUI terminals.

  • Infrastructure: Isolated Docker-in-Docker sandbox (GPU supported) and advanced ROS topic utilities.


Packages

#.. toctree:: # :maxdepth: 2 # :caption: Miscellaneous

# bob/bob-docker-network.md # bob/bob-portainer.md