Getting started with osmnxr

In osmnxr: Download, Model and Analyze 'OpenStreetMap' Street Networks

knitr::opts_chunk$set(collapse = TRUE, comment = "#>", fig.width = 6.5, fig.height = 5.5)

library(osmnxr)

What is osmnxr?

osmnxr is OSMnx for R: it downloads, models, analyzes and visualizes street networks from OpenStreetMap. The public API is tidyverse-friendly and returns sf objects; the heavy graph computation (routing, metrics, simplification) runs in a bundled Rust core.

The central object is the osm_graph: a pair of sf tables (nodes and edges) plus metadata.

A real network

The usual entry point is a place name, which downloads from OpenStreetMap:

g <- ox_graph_from_place("Olinda, Brazil", network_type = "drive")
g <- ox_simplify(g)

So this vignette runs offline, we load that exact network — the historic centre of Olinda, Brazil — from the copy bundled with the package:

g <- ox_example("olinda")
g

plot(g)

Network statistics

ox_basic_stats(g)

Routing

Snap coordinates to graph nodes, then compute the shortest path (Dijkstra, in Rust):

orig <- ox_nearest_nodes(g, x = -34.8553, y = -8.0089)
dest <- ox_nearest_nodes(g, x = -34.8505, y = -8.0125)
route <- ox_shortest_path(g, orig, dest)
length(route) # nodes along the route

route_xy <- sf::st_coordinates(g$nodes)[match(route, g$nodes$osmid), ]
plot(g, col = "grey80", lwd = 0.6)
lines(route_xy, col = "#b7410e", lwd = 3)

Where to next

• Routing and isochrones — travel time, route alternatives, service areas.
• Urban metrics — circuity, centrality, chokepoints.
• Street orientation — grid order across cities.
• Accessibility analysis — access to schools, hospitals, parks.
• Features and points of interest — download POIs and buildings.

No network needed at all? example_osm_graph() builds a synthetic grid for quick experiments.

osmnxr documentation built on Oct. 20, 2026, 1:07 a.m.