Products and Solutions
- Solutions and Technologies
  Solutions and Technologies
  
  Industry 4.0
  Building a unified digital enterprise ecosystem: data collection, digital twins, and performance management.
  
  Digital Factory
  
  Digital City
  
  Digital Infrastructure
  
  Cyber-Physical Systems
  
  Marking and Traceability
  Full-cycle mandatory labeling implementation: code application, quality validation, and integration with the Chestny ZNAK system.
  
  Industrial Vision
  Product quality control automation: external defect detection, geometry inspection, and high-speed sorting.
  
  Size and Shape Inspection
  
  Assembly Control
  
  Visual Quality Control
  
  Custom Inspection Complexes
  
  All Solutions and Technologies
  
  Digital and smart factories
  
  Industry 4.0 technologies for the transition to digital, smart, and virtual factories with flexible manufacturing and global logistics.
- Products
  
  Products
  
  Chestny Znak Marking
  A product labeling and traceability solution with Kvantron Smart software.
  
  Kvantron Smart Camera
  Industrial camera for quality inspection systems, including Smart Preform and Cap Control.
  
  Kvantron Preform Inspection Machine
  Online quality control system for PET preforms.
  
  Kvantron Smart Cap Control
  Automatic inspection of caps, closures, and sealing quality on the production line.
  
  Kvantron Smart Sorter
  Vegetable sorting system powered by computer vision.
  
  Kvantron Smart AP450 Presetter
  Offline presetting of cutting tools outside the machine.
  
  All Products
  
  Chestny ZNAK Marking
  
  End-to-end service package: measurement and analysis of the item, creation of accurate 3D models and drawings, production and verification of prototypes.
- Software
  
  Software
  
  Kvantron Smart
  Platform for computer vision and industrial inspection systems.
  
  Kvantron Smart Labeling
  Software platform for product labeling, aggregation, and traceability.
  
  All Software
Investors and Partners
Career

Digital Twin of Infrastructure Based on NVIDIA Omniverse

We create digital twins of infrastructure objects based on NVIDIA Omniverse, OpenUSD, NVIDIA Cosmos and geospatial data.

Such a digital twin can not only visualize an object, but also model its operation: the movement of people, transport, luggage, cargo, robots, special equipment, cameras, sensors and AI agents.

The solution is suitable for airports, railway stations, transport interchange hubs, ports, logistics hubs, stadiums, bridges, tunnels, and other objects with high traffic density and complex operation.

NVIDIA Omniverse is officially described as a set of libraries and microservices for developing physical AI applications, including industrial digital twins and robotic simulation.

What is a Digital Infrastructure Twin

The digital twin of infrastructure is an interactive 3D environment where buildings, territory, engineering systems, transport flows, passenger routes, cameras, sensors, equipment, and operational scenarios are brought together.

Unlike a standard 3D model, the digital twin answers not only the question “What does the object look like?” but also questions like:

How does the object operate in different modes?
Where do queues, jams, and bottlenecks form?
How do flows change during reconstruction or changes of routes?
Where is the best place to install cameras, sensors, and navigation?
How will people, transport, robots, and autonomous systems behave?
What happens in case of accident, overload, delay, or evacuation?
How can scenarios be tested before implementation at a real facility?

What Facilities Is It For

Airports	The airport's digital twin may include terminals, check-in zones, inspection and passport control, boarding gates, baggage systems, aprons, taxiways, aircraft service areas, parking lots, and transport entrances. You can model: passenger flows; queues for check-in, inspection, and passport control; baggage movement; operation of baggage handling systems; movement of ground vehicles on the apron; aircraft servicing; the work of cameras and video analytics; scenarios for delays, evacuation, and emergencies; autonomous transport and robotic systems.
Railway Stations	Station digital twin helps analyze passenger movement, platform load, transfers, queues, navigation, safety, and building operation. You can model: passenger flows; platform load; queues at ticket offices, turnstiles and screening zones; transfers between transport types; evacuation scenarios; the work of cameras and video analytics; ventilation, climate modes, and energy consumption; technical operation of infrastructure.
Transport Hubs	For TIHs, the digital twin lets you unite metro, railways, buses, taxis, private transport, parking, pedestrian routes, and commercial areas. You can check: transfer convenience; bottlenecks; loading of entrances and exits; operation of escalators and elevators; accessibility for PRM groups; partial closure scenarios; evacuation; placement of cameras and sensors.
Seaports and Container Terminals	The port's digital twin helps to model containers, cranes, stackers, tractors, vessels, rail platforms, and cargo transport. You can model: container placement; crane operations; movement of port vehicles; truck queues; loading and unloading of vessels; connection to the railway; warehouse load; emergency situations and restrictions.
Logistics Hubs and Warehouses	A digital twin of a logistics facility helps increase throughput, reduce downtime, and prepare for robotic logistics implementation. You can analyze: truck movement; operation of loading zones and gates; warehouse equipment routes; autonomous mobile robot traffic; sorting line operation; queues and bottlenecks; safety of human and vehicle movement.
Stadiums and Public Venues	For stadiums, exhibition centers, and public spaces, the digital twin helps to model people flows, security, navigation, and facility operations. You can check: visitor entry and exit; queues at checkpoints; evacuation; security allocation; zones of crowding; camera operations; emergency scenarios.

Standard Digital Twin vs Digital Twin with Physical AI

Criteria	Standard Digital Infrastructure Twin	Digital Twin Based on NVIDIA Omniverse and Physical AI
Main Goal	Object visualization, condition monitoring, data and event display.	Object operation simulation, scenario testing, AI agent training, robot, camera, and autonomous system testing.
Output Format	BIM/CAD model, 3D visualization, dashboard, facility map, dispatcher interface.	Interactive 3D environment in Omniverse/OpenUSD with physics, sensors, scenarios, and AI layer.
Human and Transport Flows	Current traffic visualization and statistics.	Future scenario modeling: queues, overloads, evacuation, route changes, autonomous logistics.
Cameras and Sensors	Used for monitoring and video analytics.	Can simulate coverage zones, blind spots, lighting, weather conditions, and AI analytics.
AI	Usually analyzes already collected data.	AI agents can be trained and tested in a virtual environment before deployment.
Robots and Autonomous Systems	Usually shown as separate elements or processes.	You can test the movement of robots, autonomous carts, UAVs, special vehicles, and service systems.
Rare Events	Difficult to analyze due to lack of real data.	Synthetic scenarios can be created: accidents, evacuations, equipment failures, weather restrictions, non-standard behavior of people or transport.
Key Value	Improved observability and easier management.	Transition from monitoring to forecasting, simulation, and digital validation of decisions.

Why Physical AI Matters for Infrastructure

Airport, station, port, or transport hub is not a static building, but a complex operational environment. People, vehicles, cargo, luggage, machinery, robots, and service teams are constantly moving within.

Physical AI enables working with such an environment: analyzing space, movement, obstacles, cameras, sensors, flows, and behavior scenarios.

In a digital twin, physical AI can be used for:

forecasting people crowds;
queue modeling;
passenger and cargo flow analysis;
safety scenario validation;
autonomous vehicle testing;
robot and special vehicle movement modeling;
preparing synthetic data for video analytics;
AI agent training for facility monitoring and management.

NVIDIA Cosmos is an open platform for physical AI with global foundation models, video processing libraries, assessment tools, and model retraining capabilities.

Notable Example: Singapore Airport Digital Twin

Why Changi Airport is a Benchmark

The Changi Airport in Singapore is one of the most technologically advanced airports in the world and a great example of infrastructure complexity. For such airports, the digital twin is particularly valued as it needs to account for passenger flows, luggage, transport, apron operations, safety, engineering systems, and facility operations.

According to open data by BuildSG, the upgrade of Changi Airport Terminal 2 used digital technologies, including virtual 3D models and specialized simulation software. These tools were applied in planning and designing the upgrade.

Note: Open sources do not confirm that the Changi digital twin was implemented specifically on NVIDIA Omniverse. Therefore, it's more accurate to refer to Changi as a reference for the complexity level of airport digitalization rather than proof of a particular platform.

How a Changi-Level Airport Digital Twin Could Look in Omniverse

Terminal Layer	Model of terminals, check-in halls, inspection zones, passport control areas, boarding gates, commercial zones, elevators, escalators, and walkways. What you can analyze: passenger flows; queues; navigation; waiting zone loads; terminal extension scenarios; the impact of flight delays on crowd density.
Baggage Layer	Model of baggage system: check-in counters, belts, sorting areas, accumulators, robotics systems, and manual handling zones. What you can check: bottlenecks; equipment failures; peak loads; baggage delivery time; robotization scenarios; the impact of delays on baggage flow.
Apron Layer	Model of the apron, aircraft parking stands, jet bridges, refuelers, catering, baggage carts, cleaning, technical services, and ground support equipment. What you can model: turnaround processes; ground handling delays; equipment movement conflicts; autonomous tractors and carts; camera and video analytics operation; emergency situations on the apron.
Transport Layer	Model of driveways, parking, taxis, buses, metro, cargo transport, and passenger logistics. What you can analyze: transport accessibility; peak flows; routing changes; overload zones; evacuation scenarios; special transport movement.
Security and Video Analytics Layer	Model of cameras, coverage zones, sensors, security posts, risk zones, and response scenarios. What you can check: camera coverage; blind spots; video analytics scenarios; rare events; synthetic training data for AI; AI agent operation in difficult situations.

Technology Foundation

NVIDIA Omniverse

Engineering 3D-environment for assembling a digital twin, physical simulation, visualization, camera modeling, sensors, robots and autonomous systems. Omniverse is used for developing physical AI applications, industrial digital twins, and robotics simulation.

OpenUSD

Unified scene structure to combine BIM, CAD, 3D models, equipment, sensors, cameras, transport, people, and movement scenarios.

NVIDIA Cosmos

Intelligent physical AI layer: scenario generation, synthetic data, future state prediction, video analysis, and AI agent training. NVIDIA Cosmos is positioned as a platform of global foundational models for physical AI.

Cesium

Geospatial foundation for sites where real territory matters: airports, ports, roads, transport hubs, industrial zones, and distributed infrastructure. Cesium for Omniverse adds 3D-geospatial features and 3D Tiles support to Omniverse.

Kvantron Smart

Applied level of computer vision: object detection, video analytics, identification, quality control, marking validation, and integration with industrial sites.

Infrastructure Twin Architecture

Level	Role in Project
GIS / Cesium	Geospatial base: territory, coordinates, terrain, roads, buildings, and outside context.
OpenUSD	Unified scene structure for combining BIM, CAD, 3D models, cameras, sensors, transport, and scenarios.
NVIDIA Omniverse	3D environment for visualization, physical simulation, sensor simulation, and working with the digital twin.
NVIDIA Cosmos	Scenario generation, synthetic data, video analysis, physical reasoning, and AI agent training.
Kvantron Smart	Applied video analytics and computer vision for infrastructure and industrial tasks.
Integrations	Connect to BIM, GIS, BMS, SCADA, IoT, cameras, dispatcher systems, and operational data.

What the Customer Gets

For Design	3D model of the facility; layout option validation; throughput analysis; checking camera and sensor placement; flow modeling; presentation materials for management and investors.
For Operation	single digital model of the facility; integration with operational data; condition monitoring; event analysis; bottleneck forecasting; support for dispatchers and operators.
For Safety	evacuation modeling; risk zone analysis; video analytics design; response scenario validation; AI agent training on synthetic scenarios.
For Robotization	robot movement modeling; autonomous cart validation; UAV testing; route analysis; interaction validation with people, transport, and infrastructure.

Digital Twin Creation Steps

1. Task Definition

Defining the project objective: design, operation, passenger flows, safety, robotization, video analytics, UAVs, logistics, or facility management.

2. Data Collection

Collecting BIM, CAD, GIS, 3D models, plans, flow diagrams, camera data, sensor data, BMS, SCADA, operational data, and facility regulations.

3. 3D Foundation Creation

Building a model of the territory, buildings, premises, roads, traffic zones, engineering objects, and key equipment.

4. Scene Assembly in Omniverse / OpenUSD

Uniting data into a single scene suitable for visualization, simulation, and future development.

5. Scenario Setting

Adding passenger, transport, cargo, service, emergency, and operational scenarios.

6. Add Physical AI

Connecting AI scenarios, synthetic data, video analytics, object behavior modeling, and AI agent testing.

7. Integration with Facility Systems

If required, linking the digital twin with cameras, sensors, BMS, SCADA, dispatcher systems, safety systems, and operational databases.

8. Handover and Support

Transferring the model, scenarios, documentation, visual materials, and ongoing digital twin support.

Data That May Be Needed

For the project you may use:

BIM/CAD models;
GIS data;
architectural plans;
flow diagrams;
passenger or cargo flow data;
3D equipment models;
facility photos and videos;
camera and sensor locations;
BMS, SCADA, IoT data;
operation regulations;
safety and evacuation scenarios.

Summary

Standard digital infrastructure twin helps you see the object, its data, and current condition.

Digital twin on NVIDIA Omniverse and Physical AI allows you to simulate, forecast, and validate facility performance in various scenarios in advance: from passenger flows and video analytics to autonomous robots, UAVs, evacuation, and emergency events.

For airports, stations, ports, logistics hubs, and large city objects this is a shift from a static model to an intelligent virtual management environment for future infrastructure.