Cheers to AI: ADAM Robot Bartender Makes Drinks at Vegas Golden Knights Game

2025-12-12 · Source: NVIDIA Blog · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems · Depth: Intermediate, quick

Summary

Richtech Robotics has deployed ADAM, an Automated Dual Arm Mixologist robot, at Las Vegas's T-Mobile Arena, serving drinks to Golden Knights fans. Developed with NVIDIA Isaac libraries, ADAM addresses hospitality labor shortages and enhances customer experience. The robot was trained in a virtual bar using NVIDIA Isaac Sim and Isaac Lab to refine skills like pouring and shaking, leveraging synthetic data for object recognition under varied conditions. ADAM operates on an NVIDIA Jetson AGX Orin edge AI platform, achieving real-time object detection and workspace calibration with less than 40 milliseconds of latency, powered by Isaac ROS 2, TAO Toolkit, and TensorRT. Richtech Robotics also introduced Dex, a mobile humanoid robot for industrial automation, running on NVIDIA Jetson Thor and trained with a mix of real and synthetic data from Isaac Sim.

Key takeaway

For robotics engineers developing autonomous systems, consider integrating NVIDIA's Isaac platform for simulation-based training and Jetson for edge deployment. Your projects can benefit from synthetic data generation to improve object recognition and achieve sub-40ms latency, crucial for dynamic environments like hospitality or industrial automation. This approach can significantly reduce development time and enhance robot adaptability.

Key insights

Robots trained in simulation with synthetic data can perform complex real-world tasks with high precision and low latency.

Principles

• Simulation accelerates robot skill refinement.
• Synthetic data improves object recognition robustness.
• Edge AI platforms enable real-time robotic perception.

Method

Robots are trained in a high-fidelity virtual environment using synthetic data for object recognition and Isaac Lab for skill refinement, then deployed on edge AI platforms for real-time operation.

In practice

• Use Isaac Sim for physically accurate robot simulations.
• Generate synthetic data to enhance perception models.
• Deploy on Jetson platforms for low-latency edge AI.

Topics

• Robotics Simulation
• Edge AI
• Humanoid Robots
• Hospitality Automation
• Industrial Automation

Best for: Machine Learning Engineer, Computer Vision Engineer, AI Engineer, Robotics Engineer, MLOps Engineer

Related on AIssential

• What’s Next in Robotics? — Physical AI, powered by NVIDIA's ecosystem, accelerates robotics development through advanced simulation and data generation.
• ABB Robotics integrates NVIDIA Omniverse for industrial AI simulation — Integrating NVIDIA Omniverse into RobotStudio enables high-accuracy virtual robot training and deployment, closing the sim-to-real gap.
• R²D²: Scaling Multimodal Robot Learning with NVIDIA Isaac Lab — Isaac Lab provides a GPU-accelerated, unified simulation framework for scalable, multimodal robot learning.
• Robots’ Holiday Wishes Come True: NVIDIA Jetson Platform Offers High-Performance Edge AI at Festive Prices — NVIDIA Jetson developer kits offer scalable edge AI compute for diverse robotics and physical AI applications.
• GTC 2026: Nvidia wants to swap robotics' data problem for a compute problem — Nvidia aims to dominate physical AI by converting data collection challenges into compute-intensive simulation problems.
• Open-Source Software Is Starting to Help Robots Think — Open-source AI is rapidly lowering barriers to advanced robotics development, accelerating innovation and accessibility.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by NVIDIA Blog.