Two out of every three people are likely to be living in cities or other urban centers by 2050, according to the United Nations, meaning about 2.5 billion people could be added to urban areas by the middle of the century. This highlights an urgent need for more sustainable urban planning and public services.
The smart traffic management market alone is projected to reach $20 billion by 2027 as cities deploy AI to accommodate growing population density and tourism-driven congestion.
Improved traffic management is just one way AI is enhancing public spaces, transit and city processes — with many physical AI technologies being showcased at the Smart City Expo World Congress (SCEWC) taking place this week in Barcelona.
At the event, NVIDIA’s expanding ecosystem of physical AI partners will be on full display — including simulation and mapping companies, software vendors, manufacturers, systems integrators, and cloud and edge providers — all exhibiting their latest smart spaces applications.
The NVIDIA Blueprint for smart city AI — announced in June at NVIDIA GTC Paris — combines digital twins with NVIDIA Omniverse libraries, synthetic data generation, AI model training with vision language models (VLMs) and video analytics AI agents using the NVIDIA Blueprint for video search and summarization (VSS).
These work together in a single, streamlined workflow, enabling rapid simulation of real-world urban conditions, vast sensor integration and real-time vision AI.
Updates to the blueprint include new NVIDIA Cosmos world foundation models and VLMs to generate photorealistic synthetic data and provide physical reasoning capabilities, as well as the latest version of the VSS blueprint — part of the NVIDIA Metropolis platform for training and building vision AI agents and applications.
Access step-by-step intelligent traffic system workflows and technical recipes in new cookbooks for NVIDIA Cosmos Predict, NVIDIA Cosmos Transfer and NVIDIA Cosmos Reason.
Learn more about how five leading companies are using these NVIDIA technologies to bring secure, scalable physical AI to more cities, including Dublin, Ireland; Ho Chi Minh City, Vietnam; and Raleigh, North Carolina.
Five NVIDIA Physical AI Partners Showcasing New Technologies at SCEWC
Urban populations are soaring, infrastructure is aging, and challenges like traffic congestion, climate resilience and real-time safety demand new levels of responsiveness and intelligence.
But integrating these video- and data-intensive systems isn’t just a matter of upgrading old infrastructure. Smart city deployments require a seamless fusion of next-generation AI, cloud and edge computing, ultrafast networking and open data platforms. A global ecosystem of partners, such as the ones highlighted below, must come together to solve these technological and real-world challenges for city-scale AI solutions.
Esri
Mapping and spatial analytics company Esri is working with NVIDIA to build an AI agent that ingests, analyzes and visualizes massive amounts of camera data to generate real-time alerts and insights in an interactive geospatial map for the City of Raleigh in North Carolina.
An interactive geospatial map of Raleigh with a video analytics AI agent processing live sensor data. Images courtesy of Esri and the City of Raleigh.
This will enable city operators to quickly respond and notify key departments to efficiently resolve issues, optimize traffic flows and improve infrastructure design, boosting overall urban mobility. Tapping into components of the NVIDIA Blueprint for smart city AI will enable the city to discover, aggregate and manipulate data to help address key resident concerns, automate streetlight timing to reduce road delays and decrease carbon emissions from cars sitting in traffic.
Milestone Systems
Milestone Systems is now introducing generative AI to its XProtect video management platform — allowing users to easily extract valuable analyses and insights from their video streams and libraries, review alerts and automatically generate reports. This functionality is powered by NVIDIA Cosmos Reason VLMs that Milestone post-trained with 75,000 hours of compliant traffic video, for each VLM, from either Europe or the U.S.
Milestone is also offering these specialized VLMs-as-a-service to developers to build innovative applications. General availability is expected later this year.
XProtect customers can easily add these new, GDPR-compliant features and add an additional layer of intelligence by contextualizing video, scheduling insights and providing on-demand reporting and automatic review of alerts. These features could reduce up to 30% of operator alarm fatigue by automating video review and filtering out false alarms. Cities like Dubuque, Iowa, and Genoa, Italy, are planning to try these new capabilities, adopting advanced video intelligence solutions to enhance traffic management.
Linker Vision
Linker Vision, a pioneer in deploying the end-to-end NVIDIA Blueprint for smart city AI, is bringing physical AI to Ho Chi Minh City and Danang in Vietnam.
Building on its success in Kaohsiung City, Taiwan, where its vision AI platform cut incident response times by up to 80%, Linker Vision, tapping into AVES Reality’s simulation-ready 3D digital twins using NVIDIA Omniverse, is now harnessing the blueprint to develop city-scale AI solutions that can simulate and monitor traffic and construction activities, helping ensure projects stay on schedule while improving urban mobility and operational efficiency.
Bentley Systems, in Collaboration With VivaCity
Ireland’s Smart Dublin organization addresses urban challenges with physical AI and is tapping into Cesium, a 3D geospatial platform from Bentley Systems, and NVIDIA Omniverse to enable real-time data visualization and analysis for micromobility — like walking, cycling and using scooters — across the city.
In addition, AI transportation technology company VivaCity is using the NVIDIA Jetson and Metropolis platforms on its AI-powered computer vision sensors, providing Dublin with highly accurate, multimodal data and insights on cyclists, motor vehicles and pedestrians. This data is used to understand traffic patterns and road user behavior, as well as to identify dangerous sites using VivaCity’s award-winning solution across the Irish capital’s roadways, improving road safety and traffic flow.
Video courtesy of Smart Dublin and Bentley Systems.
Smart Dublin is also working with partners to integrate AI agents into a digital twin to gain insights and alert city operators faster. As a first step, Smart Dublin integrated VivaCity data into a Cesium-powered digital twin to visualize and understand bike route utilization. Adding rainfall data, the city saw that poor weather had a negligible impact on the number of cyclists on the road. Future plans include adding AI features like natural language search.
In addition to the Smart Dublin project, Bentley’s AI technology Blyncsy uses NVIDIA Cosmos and Metropolis to generate synthetic data for analyzing road conditions and improving maintenance.
Deloitte
Deloitte is applying AI to automate street inspections across thousands of crosswalks for city transportation departments, aiming to protect vulnerable roadway users such as pedestrians and bikers.
NVIDIA Cosmos Transfer simulates real-world conditions and generates videos for different scenarios. Video courtesy of Deloitte.
With Cosmos Predict, Deloitte converts static intersection images into photorealistic, physically accurate videos. Then, Cosmos Transfer generates videos across a mixture of real-world scenarios, such as fog, rain, snow and low light — bringing each environment to life. Finally, Cosmos Reason helps evaluate these videos and identify potential areas of improvement, making the process insightful and dynamic.
Join NVIDIA and Partners at Smart City Expo World Congress
The NVIDIA hardware and other ecosystem partners making such smart city deployments possible include AAEON, Advantech, Aetina, Dell Technologies, HPE, OpenZeka and YUAN High Technologies.
Many of these partners will exhibit physical AI demos running on NVIDIA RTX PRO Servers, NVIDIA DGX Spark — the world’s smallest AI supercomputer — as well as NVIDIA Jetson Thor modules on the show floor.
Plus, see Akila’s physical AI technologies in the Monaco-Monte-Carlo train stations, as well as its digital twin solutions for the University Mohammed VI Polytechnic campus in Morocco.
And K2K will be demonstrating its real-time analytics solution that uses NVIDIA Cosmos Reason and the VSS blueprint to optimize roadway safety in addition to traffic and waste management.
See city-scale AI and urban innovation in action by joining NVIDIA and its partners at SCEWC and sign up to be notified when the NVIDIA Blueprint for smart city AI becomes available.
Stay up to date by following NVIDIA AI on LinkedIn, Instagram, X and Facebook.
