Overview
In the first story of this series, Highland Park used Digital Twin and AI to make building facades feel more recognisable, more useful, and better suited to everyday planning conversations. That work focused on the vertical face of the city: buildings, streetscapes, terrain, and the details that help staff and residents understand place.
The next challenge is movement.
Transportation corridors are not simple lines on a map. They are layered civic systems: pavement, frontage roads, ramps, signs, bridges, slopes, terrain transitions, access points, and driver-facing information. For a city like Highland Park, corridors such as the North Tollway Corridor and the US 75 Corridor NB are more than regional connections. They shape daily mobility, emergency access, development decisions, wayfinding, and the public's perception of the city.
Traditional 2D transportation maps can show centerlines, parcels, and right-of-way boundaries, but they rarely communicate the real experience of travelling through a corridor. A line on a map cannot easily explain how a freeway edge meets the surrounding terrain. A sign inventory table cannot show how traffic signs appear in context. A static plan view cannot help a non-technical stakeholder understand how a corridor feels at eye level.
Highland Park's second Digital Twin use case explores how AI, LiDAR, ArcGIS Pro, CityEngine, and street-level imagery can work together to transform transportation corridor data into a usable 3D environment. The goal is not to create a decorative model. The goal is to create a practical digital corridor that city staff can inspect, discuss, update, and use in everyday decision-making.
Seeing Transportation Corridors in 3D
A transportation digital twin begins with a simple but important shift: instead of treating corridors as flat road layers, it treats them as spatial systems.
For this phase, the scope focused on two major Highland Park corridors:
The work concentrated on corridor centerline extraction, freeway traffic sign locations, terrain alignment, procedural 3D corridor generation, and AI-assisted signboard texture creation. This scope gave the team a practical target: build the corridor spine, place it correctly on the terrain, enrich it with recognisable freeway signs, and prepare it for use in a 3D scene.
From a geospatial perspective, this matters because transportation corridors are alignment-sensitive. If the terrain is wrong, the road floats or sinks. If the centerline is inconsistent, the procedural model breaks. If signs are placed without a visual context, the model becomes a simplified asset inventory rather than a realistic operational view.
This is where the Esri ecosystem becomes valuable. ArcGIS Pro provides the environment for processing LiDAR and preparing terrain and corridor datasets. The 3D Basemaps workflows help turn elevation and feature data into structured 3D content. CityEngine adds procedural design logic, allowing the corridor to be generated consistently from street design rules rather than being manually modelled segment by segment. ArcGIS Online and web scenes then make the result easier to share with staff and stakeholders.
The Corridor Blueprint: From LiDAR to a Living 3D Transportation Scene
Highland Park's corridor workflow followed the same practical philosophy as the first blog: start with authoritative spatial data, process it carefully, then use AI where it adds speed and realism.
1. Building the Terrain Foundation
The workflow began with LiDAR data. Raw LiDAR provides a dense 3D record of the real-world surface, but it must be prepared before it can support a corridor model. In ArcGIS Pro, the LiDAR data was processed into usable elevation surfaces and point-cloud-derived layers.
The Digital Terrain Model became the foundation for corridor placement. This step is critical. A corridor model that is visually impressive but misaligned with the terrain cannot be trusted. By preparing the DTM first, the team created a reliable ground surface for later 3D alignment.
2. Extracting the Corridor Spine
Once the terrain was ready, the next task was to identify the transportation corridor centerlines. These centerlines act as the procedural spine of the model.
For the North Tollway Corridor and US 75 Corridor SB, the extracted centerlines provided the geometric path that CityEngine could use to construct roadway segments, corridor surfaces, and related corridor features. In practical terms, this step converted transportation geography into design-ready logic.
This is also where quality control matters. Small centerline errors can create visible issues downstream: broken segments, incorrect lane orientation, poor ramp transitions, or corridor pieces that do not sit naturally on the terrain.
3. Locating Freeway Traffic Signs
Transportation signs are small compared with the corridor, but they are large in civic meaning. They orient drivers, communicate rules, identify exits, and shape how people navigate the roadway environment.
For this phase, freeway traffic sign locations were identified for both target corridors. These locations became 3D placement points for sign structures and signboards in the corridor scene.
The aim was not only to know where a sign exists, but also to represent what the driver sees. That required a second layer of enrichment: sign-board texture extraction.
4. Designing the Corridor with CityEngine
With the terrain and centerline geometry prepared, CityEngine served as the procedural modelling environment for the corridors.
Using CityEngine's street design tools and a CGA rule file, the corridor was generated procedurally rather than hand-modelled asset by asset. This approach allowed the team to define repeatable design logic for the corridor: roadway surfaces, corridor structure, sign placements, and alignment behaviour.
The DTM was used as the terrain surface, and the generated corridor elements were aligned to that terrain. This gave the model a stronger relationship to the real landscape. Instead of hovering above a generic base map, the corridor followed the city's actual elevation context.
5. Using Street Imagery for Real-World Sign Context
Google Street imagery was used as a visual reference for the freeway corridors. The imagery helped identify the sign context and capture the signboard's appearance. This was important because generic 3D signboards can indicate that a sign exists, but they do not convey the corridor's actual condition.
By referencing street-level imagery, the model gained a recognisable transportation layer. This is the same principle used in the first blog on building facades: the more closely the digital twin resembles the real place, the easier it is for everyday users to trust and understand it.
6. Applying AI-Assisted Sign-Board Textures
The team used a custom AI web app to extract signboard textures from the street imagery. Instead of manually cropping, cleaning, and preparing every sign texture, the AI-assisted process helped isolate the board area and produce usable texture assets for the 3D sign boards.
This is a practical use of AI: not replacing GIS judgment, but accelerating a repetitive visual preparation task. Human review still matters. The AI output needs to be checked for legibility, correct sign identity, and alignment with the right 3D sign location.
Once prepared, the textures were applied to the generated signboards in the 3D corridor scene.
7. Publishing for Review and Everyday Use
The final transportation corridor model can be prepared as part of a web scene for review and sharing. This is where the workflow becomes useful beyond the technical team.
Planning staff can inspect the corridor context. Public works teams can understand sign placement and roadway relationships. Leadership can review corridor improvements in a more intuitive format. Non-GIS users can view the corridor in 3D rather than having to interpret multiple technical layers.
The result is not just a model of a road. It is a shared visual reference for how a major transportation corridor exists in the city.
The Digital Advantage: How the Corridor Twin Helps the City
Highland Park's transportation corridor twin supports a different kind of conversation. Instead of asking people to imagine the corridor from plans, profiles, and tables, it gives them a 3D environment that carries spatial context.
1. Better Corridor Planning
The model helps staff understand how corridor elements relate to terrain, nearby development, and the surrounding urban fabric. This can support discussions around capital improvements, roadway changes, signage, access, streetscape planning, and coordination with regional transportation partners.
2. More Useful Asset Context
Traffic sign locations become easier to inspect when they are represented in the corridor where they actually belong. A sign point in a database is useful, but a sign board placed in a 3D corridor with a recognisable texture tells a clearer operational story.
3. Stronger Public and Internal Communication
Transportation decisions can be difficult to explain with technical drawings alone. A corridor digital twin helps communicate what exists today and what may change tomorrow. This is especially valuable for leadership briefings, interdepartmental coordination, and public-facing discussions.
4. Faster Iteration Through Procedural Modelling
CityEngine and CGA rules make it possible to adjust corridor logic without having to rebuild every component manually. This gives the city a stronger foundation for future corridor updates, scenario planning, and model refinement.
5. Practical AI, Not AI for Show
The custom AI web app was used, which made the workflow faster and more realistic for extracting signboard textures from street imagery. This is the kind of AI that fits everyday government work because it reduces manual effort while still leaving room for human QA.
This Is What We Learned
Everyday users need context, not complexity. The final scene should help people understand the corridor quickly without requiring them to know the technical workflow behind it.
This Is What Not To Do
You May Find Useful
Closing Thought
In Series 1, Highland Park used AI to help the digital twin more closely resemble the built environment people recognise every day.
In Series 2, the same idea moves into transportation.
The corridor is no longer just a centerline. It becomes terrain- and sign-aware, and easier to understand in 3D. For a city trying to make faster, clearer, and more collaborative decisions, that shift matters.
The digital twin becomes more than a model. It becomes a common language for how the city moves.