{"id":43957,"date":"2026-03-20T14:41:00","date_gmt":"2026-03-20T14:41:00","guid":{"rendered":"http:\/\/www.prnasia.com\/story\/archive\/4915372_AE15372_0"},"modified":"2026-03-20T14:41:00","modified_gmt":"2026-03-20T14:41:00","slug":"xgrids-at-nvidia-gtc-2026-bridging-real-world-spaces-and-physical-ai-through-real2sim","status":"publish","type":"post","link":"https:\/\/transmediavictoria.net.au\/?p=43957","title":{"rendered":"XGRIDS at NVIDIA GTC 2026: Bridging Real-World Spaces and Physical AI Through Real2Sim"},"content":{"rendered":"\n

SAN JOSE, Calif.<\/span>, March 20, 2026<\/span> \/PRNewswire\/ — For robots to operate reliably in the real world, they must train in environments that accurately represent it. This is the core challenge behind Real2Sim\u2014and the focus of XGRIDS’ presence at GTC 2026 (March 16\u201319, San Jose).<\/p>\n

\n

\"\" <\/a>
<\/span><\/p>\n<\/p><\/div>\n

XGRIDS’ spatial intelligence solution now supports NVIDIA Omniverse NuRec for OpenUSD-based rendering, and appeared across multiple venues: a featured Startup Pitch, robotics demonstrations within NVIDIA’s ecosystem, and a joint showcase with Amazon Web Services (AWS).<\/p>\n

Real2Sim: From Physical Space to AI Training<\/b><\/p>\n

During Startup Pitch, XGRIDS Director Sunny Liao posed a central question:
How do you build training environments that truly reflect real-world conditions?<\/p>\n

XGRIDS’ answer is a pipeline built on real-world data. By combining LiDAR and computer vision for multimodal spatial perception with high-fidelity 3D reconstruction, physical environments can be transformed into usable world models for simulation.<\/p>\n

Compared to manual 3D modeling, this approach:<\/p>\n