This R package is an interface to the Walkalytics API that calculates the walking isochrone for a source location based on map data. To get an API key, you need to register at https://dev.walkalytics.com/signin. With the free starter account, you can make up to 100 calls a week to the API. More information (geographical availability of the service etc.) can be found here.
Update 2021-02-04: The [Walkalytics API] is no longer available.
Version 0.1.0 is on CRAN, and you can install it by:
install.packages("walkalytics")For regularly updated version (latest: 0.1.0), install from GitHub:
install.packages("devtools")
devtools::install_github("zumbov2/walkalytics")Young Albert failed the entrance examination for ETH Zurich. The 16-year-old is now to move to Aarau to complete his secondary schooling. He doesn't know the town yet and is thinking about buying a bike for his everday life. Let's help Albert decide. He doesn't want to have to walk longer than 10 minutes.
We define the blokes's most important points-of-interest and call the walkalytics API with the function isochrome_pois(). We take Albert's future address as the source location. The function pois_walktimes() processes the API response object in such a way that the estimated walking times (in seconds) between the source location and the points-of-interest are returned.
# Define Albert's points-of-interest
places <- data.frame(id = c("School", "Railway Station", "Affenkasten (Pub)", "Zurlindeninsel (River Island)"),
x = c(895737, 896297, 895620, 895840),
y = c(6006558, 6006247, 6006171, 6007080))
# Call Walkalytics API and extract walking times to points-of-interest
isochrone_pois(x = 896552, y = 6006578, epsg = 3857, pois = places, key = key) %>% pois_walktimes()
# A tibble: 4 x 4
id walktime x y
<chr> <int> <int> <int>
1 Railway Station 184 896297 6006247
2 School 348 895737 6006558
3 Zurlindeninsel (River Island) 470 895840 6007080
4 Affenkasten (Pub) 500 895620 6006171Lucky Albert. He can save his money. Read more about Albert's life here.
We can go one step further and extract high-resolution walking times. For this we use the function isochrone_esri(), which returns a response object that contains a base64-encoded gzipped Esri ASCII grid with walking times for every pixel. By using the function esri_to_sgdf(), we can convert the encoded Esri ASCII grid to an object of class SpatialGridDataFrame or we use pixel_walktimes() to directly extract walking times (in seconds) for every pixel with respect to the source location.
# Call Walkalytics API and convert response object to SpatialGridDataFrame
dt <- isochrone_esri(x = 896552, y = 6006578, epsg = 3857, key = key) %>% esri_to_sgdf()
# Call Walkalytics API and extract walking times for every pixel
dt2 <- isochrone_esri(x = 896552, y = 6006578, epsg = 3857, key = key) %>% pixel_walktimes()dt can then be displayed graphically using image()
require(viridisLite)
image(dt, col = magma(10, direction = -1))
Beside the isochrone API Walkalytics also offers the possibility of querying nearby public transportation stops. The query works in the almost same way as the previous examples.
# Call the Walkalytics pubtrans API and extract the walking times to the stations
pubtrans_ch_nearby(x = 8.528872, y = 47.382902, max_walktime = 10, key = key) %>% get_stops()
# A tibble: 7 x 8
name walktime station_category latitude longitude coordinates_type transport_category id
<chr> <dbl> <chr> <chr> <chr> <chr> <chr> <chr>
1 Zürich, Röntgenstrasse 1.30 3 8.529264 47.381932 WGS84 Bus 8591322
2 Zürich, Limmatplatz 3.40 2 8.531623 47.384600 WGS84 Bus_Tram 8591257
3 Zürich, Militär-/Langstrasse 4.40 2 8.527627 47.379600 WGS84 Bus 8591277
4 Zürich, Quellenstrasse 6.10 2 8.528753 47.386740 WGS84 Bus_Tram 8591306
5 Zürich, Kanonengasse 7.70 3 8.530306 47.378468 WGS84 Bus 8591219
6 Zürich, Museum für Gestaltung 8.50 2 8.534937 47.382121 WGS84 Bus_Tram 8591282
7 Zürich, Dammweg 9.20 2 8.526392 47.388490 WGS84 Bus_Tram 8591110Walkalytics also offers the possibility to save isochrones with freely selectable break values as PNG images (isochrone_png() %>% save_png()).
Happy testing!