Skip to main content

Chapter 2: The Robot System

Now that you understand embodied intelligence, let's look inside a robot. What makes it perceive? How does it move? What's the "glue" that coordinates all the pieces?

In this chapter, you'll explore the three essential layers of every robot:

  1. Perception (sensors that tell the robot what's happening)
  2. Actuation (motors and joints that let the robot move)
  3. Coordination (middleware that ties everything together)

You'll also take a crucial step: identifying your own hardware tier. This determines what you can build in future chapters and guides which exercises are available to you.

Duration: 4 lessons, 3 hours total Layer Breakdown: L1: 85%, L2: 15% (brief AI perspective on middleware) Hardware Tier: Tier 1 (conceptual diagrams + interactive tier selector) Prerequisites: Chapter 1 (physical AI foundations)

Learning Objectives​

By the end of this chapter, you will be able to:

  • Categorize sensor types (proprioceptive vs. exteroceptive) and understand what each type tells the robot
  • Explain actuator control (servo motors, control loops, feedback) and why open-loop control fails
  • Understand why middleware exists and what problems it solves in multi-node systems
  • Identify your hardware tier (Tier 1-4) and understand what learning paths are accessible to you
  • Recognize that robot systems are ecosystems, not monolithic units

Lessons​

Lesson 2.1: How Robots See (Sensors) (45 minutes)​

Explore the sensor ecosystem. From encoders that tell you joint position to LIDARs that map the world, sensors are your robot's eyes, ears, and proprioceptive sense.

Core Concepts:

  • Sensor types (proprioceptive: encoders, IMU; exteroceptive: LIDAR, camera, force/torque)
  • Sensor data flow and uncertainty

Lesson 2.2: How Robots Move (Actuators) (45 minutes)​

Motors aren't just "spinners." Understand servo control, feedback loops, joint limits, and why a motor's specification sheet tells you what movements are possible.

Core Concepts:

  • Actuator types (servo motors, stepper motors, brushless DC) and their properties
  • Control loops (commanded position → motor → feedback → error correction)

Lesson 2.3: The Glue (Middleware) (45 minutes)​

Why does a robot need ROS 2? Explore the problem of coordinating 10+ sensors, motors, and controllers without a unified framework.

Core Concepts:

  • The middleware problem (N sensors × M readers = complexity explosion)
  • ROS 2 as the solution (pub/sub, standard interfaces, decoupling)

Lesson 2.4: Your Hardware Tier (45 minutes)​

Take a self-assessment and identify which tier of hardware you have access to. This determines what content is available and how fast you can run exercises.

Core Concepts:

  • Hardware tier classification (Tier 1: laptop/cloud, Tier 2: GPU, Tier 3: Jetson, Tier 4: physical robot)
  • Learning path selection based on equipment

4-Layer Teaching Method​

Layer%What's Covered
L1: Manual85%Direct teaching with diagrams, sensor visualizations, motor specifications, architecture comparisons
L2: AI Collab15%Brief AI perspective on middleware design trade-offs (your first taste of AI collaboration)
L3: Intelligence0%Not yet—too early for creating reusable patterns
L4: Spec-Driven0%Not yet—specifications require coding experience

This chapter remains primarily manual foundation, but you'll get a brief introduction to AI collaboration in Lesson 2.3.

Hardware Requirements​

Minimum Tier: Tier 1 (Laptop/Browser)

TierEquipmentWhat You Can Do
1Laptop + BrowserAll lessons via cloud/browser simulation, hardware tier selector
2RTX GPU WorkstationTier 1 + local Gazebo simulation (later chapters)
3Jetson Orin KitTier 2 + real sensor data (later chapters)
4Physical Robot (Unitree Go2/G1)Tier 3 + real-world testing (capstone and Module 2+)

All Tier 1 content is fully accessible. Higher tiers enable faster execution and more advanced experiments.

Prerequisites​

  • Chapter 1 (physical AI foundations)
  • No hardware needed (all diagrams and simulators are browser-based)
  • Basic physics intuition (gravity, motion, forces—nothing formal)

Mastery Gate​

Before proceeding to Chapter 3, you should be able to:

  • Match 5 sensor types to their use cases (e.g., "I need to know the robot's orientation" → IMU)
  • Explain one actuator (e.g., servo motor) and its control mechanism
  • Answer this question: "Why is ROS 2 better than writing custom code for each sensor/motor combination?"
  • Identify your hardware tier from the Lesson 2.4 self-assessment
  • Understand that hardware tier affects content availability in Chapters 4-7

If you can do these, you're ready for Chapter 3.

Previous Chapter: ← Chapter 1: Physical AI

Next Chapter: Chapter 3: Meet ROS 2 →

Module Overview: ← Back to Module 1

Start Lesson 2.1: How Robots See →