Skip to main content

Module 1: The Robotic Nervous System - ROS 2

Decoding the Future of Humanoid Robotics​

Welcome, future architects of physical intelligence! In this module we learn the concepts and architecture of ROS 2 and how it serves as the central framework for humanoid robotics.

Learning Outcomes​

Upon completing this module, you will be able to:

  • Explain the architectural differences and advantages of ROS 2 over ROS 1.
  • Describe nodes, topics, services, and actions and when to use each.
  • Map AI agent components (perception, planning, control) onto ROS 2 architecture.
  • Outline URDF/Xacro modeling workflow and visualization with RViz2.
  • Apply debugging and best-practice approaches conceptually in a development process.

Why ROS 2 Matters​

Key points:

  • Distributed, reliable communication via DDS.
  • Improved real-time support and QoS control.
  • Language-agnostic ecosystem allowing Python/C++ interop.
  • Strong community and industry adoption.

Core Concepts (High Level)​

Nodes, Topics, Publishers/Subscribers​

Nodes are independent processes. Topics are named channels for asynchronous streaming data. Publishers send, subscribers receive.

Services vs Actions​

Services: synchronous request/response for short operations.
Actions: long-running goals with feedback and cancellation.

DDS and QoS​

DDS provides discovery and QoS (reliability, durability, deadline) that determine communication behavior.

Real-time & Safety​

ROS 2 supports real-time patterns but real hard-real-time requires OS/kernel configuration and careful system design.

Hands-on (Non-executable, Conceptual Steps)​

  1. Workspace planning: create a workspace layout with one package per functional area (perception, planning, control).
  2. Node design: list required nodes, their responsibilities, and the topics/services/actions they use.
  3. Message design: define the high-level messages and expected fields (names and semantics).
  4. Simulation test plan: design test cases to validate message flows and failure scenarios.
  5. Debug checklist: common checks (QoS mismatch, message size, stalled executors).

Note: This page intentionally contains conceptual steps only. Implementation code lives in example repos (link in Appendix → Resources) if you decide to add runnable examples later.