在 R 中使用 `mutate_at` 或 `map` 函数来计算多个位置之间的距离

Question

I have on dataframe ( controids ) with longitudes and latitudes for a set of locations:

library(tidyverse)

centroids <- 
  tribble(
    ~city, ~long, ~lat, 
    "A", -89.92702, 44.19367, 
    "B", -89.92525, 44.19654,
    "C", -89.92365, 44.19756, 
    "D", -89.91949, 44.19848, 
    "E", -89.91359, 44.19818) 

I have a second dataframe ( towns ) with longitudes and latitudes for a second set of places.

towns <- 
  tribble(
    ~town, ~long, ~lat,
    "greentown", -89.92225, 44.19727,
    "bluetown", -89.92997, 44.19899,
    "redtown", -89.91500, 44.19600)

I want to add three columns to centroids giving the straight-line distance, in kilometers, between each city and each of the three towns in towns . So the final dataframe would look like this (distances not correct, just for illustration):

output <- 
  tribble(
    ~city, ~long, ~lat, ~greentown_dist, ~bluetown_dist, ~redtown_dist,
    "A", -89.92702, 44.19367, 5.3, 2.0, 1.2,
    "B", -89.92525, 44.19654, 4.4, 2.3, 9.9,
    "C", -89.92365, 44.19756, 3.7, 5.4, 3.3,
    "D", -89.91949, 44.19848, 2.6, 3.9, 6.7,
    "E", -89.91359, 44.19818, 10.2, 2.2, 3.1)

I have to do this for a large number of towns so I'm trying to write some code that's easy to generalize. Here is what I have so far.

library(sf)

towns <- towns %>% st_as_sf(., coords=c('long', 'lat')) %>% st_geometry()

output <- 
  centroids %>% 
  st_as_sf(., coords=c('long', 'lat')) %>% 
  mutate(greentown_dist = st_distance(geometry, st_point(c( unlist(towns[1]) ))), 
         bluetown_dist = st_distance(geometry, st_point(c( unlist(towns[2]) ))), 
         redtown_dist = st_distance(geometry, st_point(c( unlist(towns[3]) ))))

I want to know if there's a way to do this using mutate_at and/or a purrr map function - one that fills in the TOWN_dist column name automatically and inputs the correct row from the town dataframe.

Answer 1

We can use map to loop over the 'towns'

centroids[paste0(c("green", "blue", "red"), "town_dist")] <- map(towns,
       ~ centroids %>% 
            st_as_sf(., coords = c('long', 'lat')) %>%
            transmute(dist = st_distance(geometry, st_point(c( unlist(.x))))))