Uber Road Geometry, Turn restrictions, Directionality & Lane connectivity Improvements(US)

Communities United States
1

Hello OpenStreetMap community,

We’re excited to announce an upcoming organized editing initiative by Uber focused on improving road network / basemap data across the United States.

As an initial phase, we plan to begin this work in New York City.

The objective of this project is to improve the accuracy and quality of the underlying road network data to support more reliable routing and navigation. Specifically, we will be focusing on:

  • Road Geometry - Creating/correcting misaligned or inaccurate road shapes against aerial imagery

  • Turn Restrictions - Adding or correcting incorrect turn restriction relations at intersections

  • Directionality - fixing incorrect one-way tagging and travel direction on ways

  • Lane Connectivity - improving lane count, lane tagging, and lane connectivity information at complex intersections and interchanges

We will begin editing after giving the community time to review this announcement and provide feedback. We expect to begin work in New York City in the upcoming week, before evaluating expansion to additional US cities.

In the spirit of transparency and collaboration, we have published full project details on the OSM wiki:
Uber Road Geometry, Turn restrictions, Directionality & Lane connectivity Improvements

The project has been registered on the Organised Editing Activities page:

Organised Editing Activities:

Uber Road Geometry, Turn restrictions, Directionality & Lane connectivity Improvements

To help illustrate the types of improvements we plan to make, below are a few representative examples from New York City. These examples are intended to show the kinds of road network issues we will review, such as inaccurate road geometry, missing or incorrect turn restrictions, incorrect one-way/directionality tagging, and incomplete lane connectivity. All edits will be verified with available aerial imagery (and, where applicable, Street-Level Imagery) before they are made.

Location Type of Edit
OpenStreetMap Missing Road segment
OpenStreetMap Missing Road segment
OpenStreetMap Missing Parking Road segment

We welcome any questions, concerns, or suggestions from the community, please reply here or reach out directly.

Thank you for your continued collaboration!

4 Likes
2

You mention

On the wiki you say you will use whatever imagery is most recent, combined with GPS tracks.

Will you be aligning solely using the GPS tracks, or will the imagery be a factor in the alignment as you mention here? How will you determine which imagery is the most recent and best aligned?

Your three examples are all missing service roads. Is the scope of this project just service roads, or all roads?

3

Hi @SherbetS

To answer your questions regarding our mapping processes and scope:

  • Imagery vs. GPS Alignment: Our internal GPS probes are primarily used as signals to identify areas that need review, as they provide highly recent data. However, we do not align roads solely based on these internal tracks. Edits are only performed when they are supported by available aerial imagery. This ensures that any changes we make can be independently validated by other community members using accessible mapping sources.

  • Scope of Road Types: While the representative examples provided happened to highlight service roads, the scope of this project is not limited to them. We will be updating all types of roads and classifications, provided there is clear evidence available to support the improvements.