Niantic Spatial and HMCI Are Building the Foundation for City of Rancho Cordova's First Digital Twin for Physical AI
For most of the last decade, a city-scale digital twin was a luxury good. Global mega cities could commission bespoke models of a city and spend years and millions of dollars building it. Most other cities were priced out.
That is no longer the case. Drone capture and modern reconstruction have replaced survey crews and hand-modeling, and the accuracy went up, not down.
Today we're announcing that Niantic Spatial is working with the Human Machine Collaboration Institute (HMCI) and the City of Rancho Cordova to build a digital twin of the city. The work is part of AiR Rancho Cordova, the AI and robotics ecosystem the City and HMCI operate together.
Rancho Cordova wants to become a working model for how a mid-sized city builds an AI and robotics economy with digital twin infrastructure, simulation-workflows, workforce development, and research partnerships that attract companies and turn experimentation into deployment. Each of those pieces has an owner. The City has put $5 million behind the initiative over three years, and since launching in January it has assembled a cohort accordingly: partners for compute and simulation, Solidigm for storage, UC Davis and Sacramento State for applied research, Folsom Lake College and the local school district for the workforce pipeline.
But simulation needs an environment to run in, research needs a shared dataset, and a company evaluating the city needs to see it before committing. Every one of those layers runs on an accurate visual model of the city, and that is the layer we own. Niantic Spatial captured and reconstructed the city's Innovation Region, roughly four square kilometers, and HMCI is already running traffic simulations against that reconstruction.
It is a portion of the city, not the whole thing. But it is real, it works, and you can watch it run.
Innovation happens when great ideas have the right environment to grow. That’s what we’re creating in Rancho Cordova. Our partnership with Niantic Spatial helps strengthen the foundation of our AI and robotics ecosystem, giving companies a place to develop and test new technologies in real-world environments. The same technology will also help the city better understand our built environment, model different scenarios, and make more informed, data-driven decisions that benefit our community. It’s another example of how we’re turning our vision into action.
Why does a city need a digital twin?
You cannot build an applied AI and robotics economy for a city nobody has accurately measured. The companies building physical AI like robotics need somewhere real to test, train, and deploy. The people planning the city need to see the consequences of a change before they commit to it. Every ambition on that list runs through the same dependency: an accurate model of the actual place.
A digital twin is a virtual representation of a real place or object, kept in sync with live data and used to monitor, simulate, and make decisions. What that definition takes for granted is the model underpinning it. For a city, that model is the hard part, as live data layered over inaccurate geometry produces answers that are confident and wrong.
From drone flight to running simulation
Over a few days this summer we flew the Innovation Region by drone and reconstructed it as a 3D model at a fidelity where you can read a building sign and measure the width of a street. Then we handed it over in a form HMCI could take straight into simulation.
The video above follows the entire path: drone capture over Rancho Cordova, city-scale reconstruction into a simulation-ready environment, and that environment running as a live traffic simulation. Capture to simulation. No hand-modeling, no multi-year procurement cycle.
What makes city-scale reconstruction hard?
There is a meaningful difference between a model that looks like a place and one that behaves like it.
Plenty of tools will generate a convincing-looking city. Far fewer produce one where the geometry holds up under measurement. That is the bar city-scale reconstruction has to clear, and it is the one we build against every time: real geometry, real coordinates, at scales that run from a single building to an entire city.
What changes when a city becomes machine-readable?
Here is the part we find genuinely exciting. Once a place exists as an accurate, machine-readable 3D model, it stops being a planning visualization and becomes the infrastructure for physical AI.
City staff can walk stakeholders through a proposed change before breaking ground. Planners can test traffic scenarios without closing a lane. A robotics company deciding where to deploy can prove its system works in Rancho Cordova before committing to it.
The city stops being something you describe in a meeting. It becomes something you can run and interrogate.
The goal is the whole city
Next, with HMCI and the City, the ambition is the rest of Rancho Cordova and richer simulations as the model grows to meet them.
There are several hundred cities in America that look a lot like Rancho Cordova. Cities with real ambition, real constraints, and no realistic path to a bespoke twin. We think many of them are going to want what this city now has.
At HMCI, we have always believed that the path to AGI will require more than building a better digital brain. We must also build machines that can act, create an accurate understanding of the physical world, and connect all three. This digital twin is an important step toward that larger vision. Niantic Spatial is making Rancho Cordova machine-readable, while Rancho Cordova is giving us something equally rare: a city willing to become a real-world laboratory where researchers, companies, and the community can safely build, test, and shape the future of physical AI together.
Rancho Cordova calls what we built a digital twin, and that is exactly what it is. But it is also something less familiar. Every reconstruction we produce is a real-world model: a machine-readable account of an actual place, built from real geometry rather than generated from a prompt. Physical AI has to run on something, and generated environments cannot tell a robot what is truly there.
We are building that foundation.
About the partners
The Human Machine Collaboration Institute (HMCI) is a research, education, and advising organization advancing how humans and machines work together. Through partnerships with industry, academia, and government, HMCI builds AI ecosystems that accelerate innovation, workforce development, and community transformation. Learn more at hmci.ai.
The City of Rancho Cordova incorporated on July 1, 2003, becoming the 478th city in California. Since that time, Rancho Cordova is one of the top five fastest growing cities in California—an emerging urban center, with a small-town feel. Over 3,500 businesses employ a workforce of 65,000+, making Rancho Cordova one of the largest employment centers in the Greater Sacramento region, as well as home to the first city-driven AI & Robotics Ecosystem in the nation. The city’s over 85,000 residents enjoy a beautiful, six-mile stretch of the American River, a burgeoning arts scene, 26 miles of bike and pedestrian trails, 70 acres of creeks and tributaries, and over 100 (mostly free) events. Whether you call Rancho Cordova an All-America City, Playful City USA, Tree City USA, or Your City, its neighborhoods and business districts reflect innovation, opportunity, diversity, and community. Learn more at cityofranchocordova.org.
Niantic Spatial builds foundation models for physical AI, giving robots, AI agents, and people an accurate, shared understanding of physical spaces. Its reconstruction technology captures environments with geometric accuracy and fine detail from any standard camera, and its Visual Positioning System delivers precise positioning almost anywhere in the world. Learn more at nianticspatial.com.