The heavy lifting in manufacturing, factory logistics, and robotics is getting assistance from real-time AI, according to NVIDIA CEO Jensen Huang’s keynote at जीटीसी एक्सएनएनएक्स.

A simulation-first approach is paving the way for the next phase of automation in these industries that often involve bulky products, expensive equipment, collaborative robotic environments, and logistically-complex facilities.

Huang demonstrated how developers could use digital twins to develop, test, and refine their large-scale, real-time AI entirely in simulation before deploying it in industrial infrastructure—saving significant time and cost. NVIDIA’s Omniverse, Metropolis, Isaac, and cuOpt platforms interact in “AI gyms” where developers can train AI agents to help robots and humans navigate unpredictable or complex situations.

In a demo, a 100,000-square-foot warehouse’s digital twin built with Omniverse operated as a simulation environment. It included dozens of digital workers, multiple autonomous mobile robots (AMRs) running NVIDIA Isaac’s multi-sensor stack, vision AI agents, and sensors. Metropolis created a centralised occupancy map by fusing data from 100 simulated camera streams to inform optimal AMR routes calculated by cuOpt’s complex routing optimisation AI.

This all happened in real-time while Isaac Mission Control coordinated the AMR fleet using cuOpt’s mapping and routing data. When an incident blocked an AMR’s path, Metropolis updated the occupancy grid, cuOpt planned a new optimal route, and the AMR responded accordingly to minimise downtime.

Using Metropolis vision models and the Visual Insight Agent framework, developers can build AI agents to help operations answer questions like “What happened in aisle three?” with insights from video analysis. These visual AI agents will help industries extract actionable insights from video using natural language.

The AI capabilities demonstrated are enhanced through continuous simulation training and deployed as modular NVIDIA inference microservices, driving the next phase of industrial automation powered by real-time AI.

(फोटो द्वारा चटर स्नैप)

इन्हें भी देखें: 2030 तक यूके के आसमान में उड़ने वाली टैक्सियाँ और डिलीवरी ड्रोन आने की तैयारी है

उद्योग जगत के नेताओं से IoT के बारे में जानना चाहते हैं? चेक आउट IoT टेक एक्सपो एम्स्टर्डम, कैलिफ़ोर्निया और लंदन में हो रहा है। व्यापक कार्यक्रम अन्य प्रमुख कार्यक्रमों के साथ सह-स्थित है साइबर सुरक्षा और क्लाउड एक्सपो, एआई और बिग डेटा एक्सपो, एज कंप्यूटिंग एक्सपो, तथा डिजिटल परिवर्तन सप्ताह.

TechForge द्वारा संचालित अन्य आगामी उद्यम प्रौद्योगिकी कार्यक्रमों और वेबिनार का अन्वेषण करें यहाँ उत्पन्न करें.

टैग: ai, कृत्रिम बुद्धिमत्ता, स्वचालन, cuopt, डिजिटल ट्विन, औद्योगिक स्वचालन, IoT, isaac, राजधानी, NVIDIA, Omniverse