Skip to content

Latest commit

 

History

History
117 lines (91 loc) · 3.79 KB

Osm-Data.md

File metadata and controls

117 lines (91 loc) · 3.79 KB
Raw

Introduction

Open Street Map is a powerful and useful open geographical data source that we can use in our simulation.

Open Street Map contains data about many features : roads, buildings, cycleways, public transports, centers of interest and activities... A model could potential use one or several of all those features.

SmartGov provides some utilities to pre-process and import OSM data directly in our simulation.

Preprocess data

OSM file

This first thing that we need in as XML .osm file, that represents a region.

Many options exist to download OSM data. However, GEOFABRIK is an useful website that allows you to download regularly updated .osm files by regions, so that they are not too heavy to apply further processes.

For the purpose of this example, we will be working on the Rhone-Alpes region.

When using GEOFABRIK, you need to download the ".osm.bz2" option to download your data.

Extract a zone

Because of the huge amount of data contained even at the region scale, the first thing that we might want to do is to crop the downloaded file, at the scale of a city for example.

In this context, Osmosis is an useful tool that will help us to crop the data and apply a first filter on OSM items.

Start by checking the Osmosis installation section, and install it on your system.

From there, a single command line will extract the required data for us :

bzcat rhone-alpes-latest.osm.bz2 | osmosis --read-xml enableDateParsing=no file=- --bounding-box top=46.146 left=4.4 bottom=45.43 right=5.46 --tf accept-ways highway=* --used-node --write-xml lyon.osm

where :

  • rhone-alpes-latest.osm.bz2 is the file downloaded from GEOFABRIK
  • lyon.osm is our output filtered data
  • top, left, bottom, right represents the bounding box of the area we want to extract, in latitude / longitude
  • --tf accept-ways highway=* --used-node is a tag filter that allows us to clean our data before other processes : here, we only keep OSM highways and nodes that belong to them. Do not use this if you want to keep other features, such as buildings. Other examples are available in the Osmosis documentation.

Convert to JSON data

In order to improve interoperability and genericity among our simulation inputs and outputs, our simulation doesn't directly take a .osm file as input, but should instead "convert it" to JSON files (a nodes and a ways file), eventually after applying more filtering.

Nodes file

The nodes file should have the following general form :

[ {
	"id": "1",
	"tags": {
		"key": "value",
		...
		},
	"lat": 0.0,
	"lon": 0.0
	},
	...
]

Ways file

The ways file should have the following general form :

[ {
  "id" : "1",
  "tags" : {
    "name" : "road name",
    "highway" : "highway type",
    "oneway" : "no/yes/-1/...",
	"other_tag": "other_value",
	...
  },
  "nodeRefs" : [ "1", "2", "3", ... ]
  },
  ...
]
  • The name tag is optional but useful
  • The highway tag should always be specified
  • The oneway tag is optional : no value means that the road is not oneway.

Notice that those format are easily human readable (and editable), could be easily imported in a web application for example, or being generated using other methods.

Conversion using the OsmParser

In order to convert our .osm file to those JSON files, a side project, called the SmartGovOsmParser has been implemented.