Skip to main content

Workstation Requirements

The Demands of Physical AI​

This course is technically demanding, sitting at the intersection of three heavy computational loads:

  • Physics Simulation (Isaac Sim/Gazebo): Simulating the real world is computationally expensive.
  • Visual Perception (SLAM/Computer Vision): Enabling robots to see and understand their environment requires significant processing power.
  • Generative AI (LLMs/VLA): Running large language and vision-language-action models is resource-intensive.

Because our capstone project involves a "Simulated Humanoid," the primary investment must be in High-Performance Workstations. However, to truly fulfill the promise of "Physical AI," you will also need Edge Computing Kits and potentially specific robot hardware.

The "Digital Twin" Workstation (Required per Student)​

This is the most critical component of your setup. NVIDIA Isaac Sim, an Omniverse application, is at the heart of our simulation work, and it requires a GPU with RTX (Ray Tracing) capabilities.

Important

Standard laptops, including MacBooks and non-RTX Windows machines, will not be sufficient for this course.

GPU (The Bottleneck)​

  • Required: NVIDIA RTX 4070 Ti (12GB VRAM) or higher.
  • Why: You need high VRAM to load the USD (Universal Scene Description) assets for the robot and environment, and to run the VLA (Vision-Language-Action) models simultaneously.
  • Ideal: An RTX 3090 or 4090 with 24GB of VRAM will provide a smoother "Sim-to-Real" training experience.

CPU​

  • Required: Intel Core i7 (13th Gen+) or AMD Ryzen 9.
  • Why: Physics calculations (Rigid Body Dynamics) in Gazebo and Isaac Sim are CPU-intensive.

RAM​

  • Required: 64 GB DDR5.
  • Minimum: 32 GB is the absolute minimum, but you will likely experience crashes during complex scene rendering.

Operating System​

  • Required: Ubuntu 22.04 LTS.
  • Note: While Isaac Sim can run on Windows, ROS 2 (Humble/Iron) is native to Linux. A dual-boot setup or a dedicated Linux machine is mandatory for a friction-free experience.