R: divide a given Box into X amount of new boxes with given diagonal

I have a Bbox of Germany with coordinates (lon,lat):

g_bbox<-c(xmin=5.98865807458, ymin=47.3024876979, xmax=15.0169958839, ymax=54.983104153)

I would like to divide the bbox in (x) amount of bboxes with a diagonal of 50 km each.

The length of diagonal can vary a little (+/-10km).

The Resulting BBoxes should not overlap.

The amount of BBoxes (x) should be the maximum number of BBoxes that fit into the g_bbox

The result I would like to have is in best case a data frame with xmin,ymin,xmax,ymax columns.

How would I go about it?

I would suggest the following solution.

Load packages

library(sf)
#> Linking to GEOS 3.9.0, GDAL 3.2.1, PROJ 7.2.1

Load data

g_bbox <- st_bbox(c(xmin = 5.98865807458, ymin = 47.3024876979, xmax = 15.0169958839, ymax = 54.983104153), crs = 4326)

Convert bbox to polygon with projected coords

g_poly <- st_as_sfc(g_bbox) %>%
  st_transform(32632)

Plot

mapview::mapview(g_poly)

Create the grid

g_grid <- st_make_grid(g_poly, cellsize = 50000 / sqrt(2))

Check one diagonal

sqrt(
  (st_bbox(g_grid[[1]])[3] - st_bbox(g_grid[[1]])[1]) ^ 2 # x axis
  +
  (st_bbox(g_grid[[1]])[4] - st_bbox(g_grid[[1]])[2]) ^ 2 # y axis
)
#>  xmax 
#> 50000

Plot

mapview::mapview(g_grid)

Estimate all bboxes

head(do.call("rbind", lapply(g_grid %>% st_transform(4326), st_bbox)))
#>          xmin     ymin     xmax     ymax
#> [1,] 5.970445 47.30249 6.455849 47.63163
#> [2,] 6.440454 47.31382 6.923301 47.64117
#> [3,] 6.910729 47.32325 7.390967 47.64877
#> [4,] 7.381223 47.33077 7.858798 47.65445
#> [5,] 7.851885 47.33638 8.326746 47.65819
#> [6,] 8.322667 47.34009 8.794764 47.66001

^{Created on 2021-06-30 by the reprex package (v2.0.0)}