🛠️ Hardware and Lab Requirements

This course is technically demanding, sitting at the intersection of three heavy computational loads: Physics Simulation, Visual Perception (SLAM/Computer Vision), and Generative AI (LLMs/VLA).

Critical Requirement

NVIDIA RTX GPU Required. Standard MacBooks or non-RTX laptops will not work for the Isaac Sim portion of this course.

1. The "Digital Twin" Workstation

This is your primary development machine.

GPU (Critical)

NVIDIA RTX 4070 Ti (12GB+) or higher.

CPU

Intel Core i7 (13th Gen+) or Ryzen 9.

RAM

64 GB DDR5 (32GB min).

OS

Ubuntu 22.04 LTS (Native).

2. The "Physical AI" Edge Kit

These components form the "nervous system" of your robot.

Component	Recommended Model	Why?
The Brain 🧠	NVIDIA Jetson Orin Nano (8GB)	Runs the ROS 2 inference stack on the robot.
The Eyes 👀	Intel RealSense D435i	Provides RGB + Depth data for 3D perception.
The Ears 👂	ReSpeaker USB Mic Array	Enables voice commands via Whisper.

💰 Budget Option: The Economy Kit (~$700)

• NVIDIA Jetson Orin Nano Super Dev Kit: ~$249
• Intel RealSense D435i: ~$349
• ReSpeaker USB Mic: ~$69
• SD Card & Wires: ~$30

3. The Robot Lab (Actuators)

Choosing the right physical body for your AI.

Option A: The "Proxy" (Best Value)

• Robot: Unitree Go2 Edu (~$1,800)
• Pros: Excellent ROS 2 support, durable.
• Cons: Quadruped (dog), not humanoid.

Option B: The "Premium" Humanoid

• Robot: Unitree G1 Humanoid (~$16,000)
• Why: Full bipedal dynamics, capable of complex manipulation tasks.

---

4. Infrastructure Summary

Your lab ecosystem connects as follows:

graph TD
    A[Sim Workstation (RTX 4090)] -->|Trains Model| B(Hugging Face Hub)
    B -->|Downloads Weights| C[Edge Brain (Jetson Orin)]
    D[Sensors (RealSense)] -->|RGB-D Data| C
    C -->|Motor Commands| E[Actuator (Robot Body)]