Overview:

  • Google describes itself as a commercial party in map services.
  • Road network comes from Government agencies, with crowdsourced edit.
  • Traffic directions updates come from crowdsourced correction.
  • Map data is private, but offers Map APIs.
  • Google may not disclose its routing algorithm.

Mapping Technology

Web Map Tiles

Map tiles are 256 x 256 pixel panels treated as display units.

  1. Projection: EPSG:3857 WGS84 / Pseudo Mercator (Web Mercator).
    1. Coordinate systems: Latitude and Longitude (WGS84 datum);
    2. Map extent: symmetrically truncated near poles to fit in a square;
  2. World coordinates: 0-256, floating point, origin at upper left corner (~85˚N, 180˚);
  3. Map at certain zoom level:
    • Tile coordinates (0-2^zoomLevel-1, integer)
    • Pixel Coordinates (0-256*2^zoomLevel-1, integer)

Zoom level is the level at which the base tile (256 x 256 pixels) is recursively refined. The zoom level of the base tile is 0. Maximum zoom level of vector map is 21.

  pixelCoordinate = worldCoordinate * 2^zoomLevel

Open Location Code (Plus Code)

Open Location Code is a base-20 numbering system for WGS84 coordinates, with pattern .(yx)(yx)(yx)(yx)+(yx)b and digits 23456789CFGHJMPQRVWX. The lat-lng plane is first divided into a 20-degree grid (918), and successively subdivided (2020) for four times; further digits may be used, each coding a lat-lng block (5*4); digits increase northward and eastward. At ten digits, the north-south precision is about 14 meters (x-y precision ratio is $\cos \varphi$).

Map Data

  • Non-road geometry: political/administrative boundaries; natural features (mountains, water bodies); parks and protected areas;
  • Road network: road center line and lanes, facilities (bridge, tunnel); pedestrian way; cycling path;
  • Directions: speed limit, one-way, traffic & parking restrictions; street names, addresses;
  • Traffic: typical traffic; live traffic and traffic incidents;
  • Transit: transit routes, schedules, fares and real-time feed;
  • Digital elevation models (DEM): Terrain (above sea level); Bathymetry (below sea level);
  • City models: 3D building models with accurate geometry and facade; 2.5D building footprints (height, base geometry, use type and address)
  • Earth imagery: satellite; aerial (45° imagery); historical imagery (TIME's Timelapse project);
  • Google Street View (explore places, road restrictions);
  • Point of interest: hospitals, tourist attractions, schools, government buildings, local businesses;

Road network directions data can be acquired and confirmed by reading street signs from Street View imagery. Roadway details like the size of the road could be recognized by (Satellite/Aerial) overhead imagery.

Google Traffic provides real-time traffic conditions on major roads and highways in over 50 countries, using technology from ZipDash. Live traffic data are crowd-sourced from Android phones which act as an army of car-mounted traffic monitors. The iPhone Maps application, however, does not support traffic crowdsourcing. Just as with the highways, the colors (of arterial roads) correspond to the speed of traffic (relative to the speed limit of the road): green is free sailing, yellow is medium congestion, red is heavy congestion, and dark red is stop-and-go traffic.

Google I/O 2013 - Fireside Chat with the Google Maps Team:

(18:39) Speed limit data turns out to be really difficult to get canonical sources for. And the only way, I think, it can be done is by driving all the roads and picking up images.

(19:20) And so actually creating a model of how fast vehicles actually travel and whether or not they're over the speed limit or under the speed limit is actually very, very difficult, because it turns out that the data is just very, very noisy.

According to Richard Russell on Quora, Google Maps' estimated time of arrival (ETA) are based on a variety of things, depending on the data available in a particular area. These things range from official speed limits and recommended speeds, likely speeds by road types, historical average speed data over certain time periods (sometimes just averages, sometimes at particular times of day), actual travel times from previous users, and real-time traffic information.

Google Transit is a public transportation planning tool which integrates transit stop, route, schedule, and fare information. Google Transit resources: Google Developers; General Transit Feed Specification (GTFS); list of public GTFS feeds; transit agencies on GTFS Data Exchange.

Product Timeline

Early versions of Google Maps provided information to users about how long it would take to travel a particular road in heavy traffic conditions.

Data Source

Google Maps Content Partners

Google Maps Content Partners/Providers are organizations that contribute authoritative map data to Google. Selected providers are identified in a credits line at the base of the user interface. Under the terms of Google's content license agreement with content partners, Google agrees not to provide partners' content to third parties in a raw, unmodified form.

List of Content Partners:

  1. Governments and the Agencies
  2. Commercial Entities
  3. Non-profit Organizations
  4. Educational Institutions

Crowdsourced User Edit

Map updates are partly based on user reports via "Report a Problem" and Google Map Maker.

  • Parcel Boundaries: residential neighborhoods, shopping centers
  • Road Network details: new highway on-ramp, new traffic light, turn restrictions, pedestrian only, renaming, geocoded addresses.
  • 360º Photo Sphere

Project Ground Truth

Google Maps' Project Ground Truth keeps Google Maps as up to date and accurate as possible for over 50 countries by:

  • reviewing crowdsourced user edits;
  • using satellite and aerial images to trace roads, bodies of water, park boundaries and more;
  • scoping changes (road closure, housing/business development, name change) from governments, organizations and "crystal reports" (scrapping social media and news sites);
  • using Street View images to determine lane directions and pinpoint businesses;

Google I/O 2013 - Project Ground Truth: Accurate Maps Via Algorithms and Elbow Grease showed some under-the-hood details of the project. Google Maps' US road network started from TIGER/Line Shapefiles, but have dramatically improved upon it. Computer vision algorithm are used to extract addresses, one-way streets, bicycle lanes, types of roadways, and business logos from Street View, using their internal mapping tool Atlas (GUI). The Ground Truth project hired and trained many operators for each country maintained using Atlas, which is quite expensive in all aspects.

Google Maps APIs

Selected Google Maps APIs: 1. Google Maps JavaScript API v3 (object-oriented programming); 1. Google Maps API Web Services * Time Zone API (time zone, time offset) * Roads API (identify roads, posted speed limit) 1. Geolocation API (Get a location and accuracy radius based on information about cell towers and WiFi nodes that a mobile client can detect.) 1. Tracks API (track, analyze, and visualize asset movement)

Pricing: 1. Free up to 25k map loads per day; 2. Purchasing additional quota (https://developers.google.com/maps/faq#usage_pricing); 3. Purchasing a Google Maps API for Work license (minimum $10k/year);

JavaScript API v3

  1. Loading Data:
    • Google Maps Data layer (GeoJSON)
    • KML layer (KML, GeoRSS)
    • Maps Engine layers (Imports from Google Maps Engine)
    • Fusion Tables layer (Google Fusion Tables)
  2. Drawing: features, info window;
  3. Layers:
    • Traffic layer: Current traffic, Typical traffic per half hour; Speed, disruptions (accidents, road work);
    • Transit layer: transit modes; route, schedule, fare, vehicle capacity.
    • Bicycling layer
  4. Services/Analytics:
    • Geocoding and Reverse Geocoding
    • Driving Directions: Without traffic, Avoid highways/tolls/ferries.
    • Transit Directions: Best route (minimal time) , Fewer transfers, Less walking; (use trails, avoid hills;)
    • Distance Matrix: travel distance and travel time for multiple origins and destinations;
    • Geometry Library
    • Elevation
    • Places Library: establishments, geographic locations, prominent points of interest;
    • Maximum Zoom Imagery
    • Street View
  5. Visualization Library: Heat map layer

Google Maps Engine API: store geographic data, visualize as overlay. [To be deprecated on Jan 29, 2016; An alternative may be Google My Maps]

Google Earth Engine: satellite imagery of the Earth over 40 years. Projects that used Google Earth Engine: deforestation, land cover change, urban sprawl; earthquake & flood risk modeling; drought monitoring, agricultural water consumption estimation, epidemic disease risk modeling; Google I/O 2014: Saving trees and saving lives with petapixel-scale computing

Directions APIs

Without traffic: DirectionsResult with route and duration in response. Available at Google Maps JavaScript API v3.

Live traffic: DirectionsRequest:durationInTraffic specifies whether the DirectionsLeg result should include a duration that takes into account current traffic conditions. This feature is only available for Google Maps API for Work customers. The time in current traffic will only be returned if traffic information is available in the requested area.

Typical traffic: Google Maps only provide typical traffic layer in half-hour intervals, no directions.

Historic traffic: Google Maps API for Work claimed to provide historic traffic data, which doesn't seem to be currently supported.


🏷 Category=Geographic Information System