M2M Tech Showcases At-Scale Physical AI Architecture at NVIDIA GTC as DDN Partner for Omniverse Integration

Live robotics, Edge AI, and AI Factory demonstrations highlight a unified Edge-to-Data-to-Simulation stack for mission-critical infrastructure

• Unified Physical AI architecture integrating deterministic Edge execution, Digital twin simulation, and DDN's AI data infrastructure layer
• Three live demonstrations across the NVIDIA Omniverse tour bus and the DDN booth experience at GTC
• Architecture designed for Defense, Healthcare, Transportation, Energy, and Advanced manufacturing at national scale

San Jose, California--(Newsfile Corp. - March 16, 2026) - M2M Tech today announced its participation at NVIDIA GTC, where it will showcase a unified Physical AI architecture built on NVIDIA Omniverse and integrated with DDN's world leading AI and data intelligence solutions. The showcase demonstrates how Edge systems and robotics can operate in a closed-loop lifecycle - capture, validate, learn, and redeploy - using simulation, Data intelligence, and deterministic Edge execution.

What M2M is Demonstrating at GTC

At the core is a working reference architecture that connects:

• M2M Edge AI (MEA) platform (execution and control at the Edge)
• NVIDIA Omniverse (simulation and Digital twins)
• DDN AI Data platform (high-performance AI Data infrastructure and system of record)

Together, this stack supports end-to-end Physical AI: Edge data capture and filtering, evidence-grade storage and retrieval, simulation-first policy and model validation, and controlled redeployment back to the field.

Live Demonstrations

1) VR-to-Omniverse Robotics Interaction and Replay

• An immersive VR experience in NVIDIA Isaac Sim where users interact with a virtual robot arm in a timed, pattern-based task to demonstrate intuitive human-in-the-loop control.
• VR inputs drive real-time robot-arm behavior inside the Omniverse simulation, showcasing low-latency interaction and synchronized state within the digital environment.
• Session events and telemetry are captured and promoted into DDN's data layer for indexing, replay, and post-session analysis.

2) Edge AI Inference in Low-Connectivity Environments

• Live Robotics and safety-focused demonstrations where MEA performs local inference without reliance on centralized cloud connectivity.
• High-value events are filtered at the edge and stored as structured evidence for search, replay, and compliance workflows.
• Demonstrates resilient operation for remote sites and constrained networks.

3) AI Factory Architecture for Multi-Site Infrastructure

Schreibe Deinen Kommentar