DeeperML

Nvidia is significantly advancing its role in humanoid robotics through its Jetson Thor platform, aiming to replicate its AI chip success. CEO Jensen Huang has identified humanoid robots as a key area for large-scale production. To further this goal, Nvidia is releasing Isaac GR00T Blueprint which helps with generating training data for humanoid robots. This blueprint leverages workflows for synthetic data creation and simulation, and includes tools like the GR00T-Teleop, GR00T-Mimic and GR00T-Gen to generate large training datasets for robots to learn imitation skills. The company is also releasing a collection of robot foundation models and data pipelines to accelerate development.

Foxconn is partnering with Nvidia to develop humanoid robot services, integrating Nvidia's advanced software and hardware in Kaohsiung City, Taiwan. This partnership marks a significant move for Foxconn to diversify beyond contract manufacturing in electronics, with plans to collaborate on smart city projects. Additionally, Taiwan is building a supercomputer based on NVIDIA’s Blackwell architecture, further emphasizing the island's commitment to AI advancement and robotic capabilities. NVIDIA's Jetson Thor computing system, debuting in early 2025, will be central to this endeavor. The company's actions highlight a broader trend towards embodied AI.

ImgSrc: img.digitimes.c

References :

• blogs.nvidia.com: Over the next two decades, the market for humanoid robots is expected to reach $38 billion.
• analyticsindiamag.com: Foxconn, NVIDIA Partner to Develop Humanoid Robot Service
• www.digitimes.com: Nvidia's robot computing platform Jetson Thor aims to replicate AI chip success story

Classification:

• HashTags: #NvidiaAI #HumanoidRobots #Robotics
• Company: Nvidia
• Target: Humanoid Robotics Industry
• Product: Jetson Thor
• Feature: AI Robotics Platform
• Type: AI
• Severity: Informative