OpenStreetMap Charging Station Completeness (CSC)

A little hobby project for carrying out periodic analysis of the completeness of tags related to electric vehicle charging stations on OpenStreetMap. Also contains an API and associated web UI for making summary queries based on location

Local setup

Docker is recommended.

Build and launch everything. It will take a while to build the csc-process service because it has to compile osm2pgsql)

docker compose up -d

Load the database with content. Everything is currently hardcoded so you don't need to change anything to get it working. By default, it will process the OSM data for the German city of Karlsruhe and surrounding area. You can change this by modifying the PBF_DOWNLOAD_LINK env var in the csc-process service in ./docker-compose.yaml. Just keep in mind that larger OSM PBF files means longer processing times.

docker compose run csc-process update

Once finished, you should have a table in your PostGIS database accessible from the csc-prod-db service, by default mapped to port 5432. The relevant statistical information about tag completeness is found in the table called cs_completeness.

Every time you call this command, a new set of charging station points will be processed in csc-process-db and then piped into csc-prod-db with a timestamp. This allows you to take periodic readings and see trends over time.

Use

API

The API has three endpoints and is accessible by default on port 8083.

/status - Sanity check endpoint, a GET request just returns 200 to indicate that the server is online.

/completeness - Analyze the completeness of charging stations via a POST request.

Specify a study area with the "studyArea" key mapping to a valid GeoJSON to filter the results to a specific area. Note that this GeoJSON must contain only one feature and it must be a Polygon or a MultiPolygon. Specify a time with the "timestamp" key mapping to an ISO-compliant timestamp to only retrieve the set of results which were current in the database at that timestamp.

Both of these keys are optional. Omitting "studyArea" will simply summarize the entire database. Omitting "timestamp" will only summarize the most recent version of charging stations.

{
  "studyArea":{
    "type":"FeatureCollection",
    "name":"ka-csc",
    "crs":{
      "type":"name",
      "properties":{
        "name":"urn:ogc:def:crs:OGC:1.3:CRS84"
      }
    },
    "features":[
      {
        "type":"Feature",
        "properties":{
          
        },
        "geometry":{
          "type":"Polygon",
          "coordinates":[
            [
              [
                8.393922325496273,
                48.991503115729685
              ],
              [
                8.394031307926372,
                49.029864931124415
              ],
              [
                8.448740487835906,
                49.029864931124415
              ],
              [
                8.448740487835906,
                48.986380941515044
              ],
              [
                8.448740487835906,
                48.986380941515044
              ],
              [
                8.393922325496273,
                48.991503115729685
              ]
            ]
          ]
        }
      }
    ]
  },
  "timestamp":"2023-09-30 00:00:00"
}

/completeness_interval - Analyze the completeness of charging stations over time via a POST request.

Specify a study area with the "studyArea" key mapping to a valid GeoJSON to filter the results to a specific area. Note that this GeoJSON must contain only one feature and it must be a Polygon or a MultiPolygon. Specify a time interval with the "timestamps" key as follows:

{
  "start": <ISO-compliant timestamp>,
  "end": <ISO-compliant timestamp>
}

Both of these keys are optional. Omitting "studyArea" will simply summarize the entire database. Omitting "timestamps" will only summarize all versions of the database.

In contrast to /completeness, this endpoint returns tag completeness results as arrays, and also returns a "timestamps" key which corresponds to the timestamp for each array value in each tag result.

{
  "studyArea":{
    "type":"FeatureCollection",
    "name":"ka-csc",
    "crs":{
      "type":"name",
      "properties":{
        "name":"urn:ogc:def:crs:OGC:1.3:CRS84"
      }
    },
    "features":[
      {
        "type":"Feature",
        "properties":{
          
        },
        "geometry":{
          "type":"Polygon",
          "coordinates":[
            [
              [
                8.393922325496273,
                48.991503115729685
              ],
              [
                8.394031307926372,
                49.029864931124415
              ],
              [
                8.448740487835906,
                49.029864931124415
              ],
              [
                8.448740487835906,
                48.986380941515044
              ],
              [
                8.448740487835906,
                48.986380941515044
              ],
              [
                8.393922325496273,
                48.991503115729685
              ]
            ]
          ]
        }
      }
    ]
  },
  "timestamps":{
    "start":"2023-09-30 00:00:00",
    "end":"2023-10-01 00:00:00"
  }
}

UI

A web-based user interface can be found by default at port 8082. It is based on Vue 3. You can draw a polygon on an OpenLayers map and submit it as a request to the API, and it will return the result based on your input.

Contributing

Just go for it. This project is not large enough, nor do I have the time or energy to warrant official contribution guidelines. If you like the project and have an idea, make an issue. If you want to code directly on it, make a fork and a PR. I'm happy for any support.

Credits

• Isaac Boates - Started the project
• Roman Karavia - Helped a lot at the Geofabrik OSM Hackathon in 09.2023, mostly with properly configuring the UI and getting it to communicate with the API