I have 2 years worth of GPS data of 2 different groups across a 400ha site, with each distinct ‘name’ (150) belonging to either group ‘a’ or ‘b’, each name has different amounts of GPS data, and each group has different numbers of individuals.
I have the mean coordinates of the distinct individuals who each belong in either group a or b. I have 1 df with 4 columns: name, group, mean lat, mean long.

1: I am trying to find any clusters within each group (to identify any herds) within a and within b.

2: I would also like to see if there are any clusters between group a and b to analyse if there are any interactions there. Cluster size and distance is unknown.

Will hierarchical analysis, dbscan, or K-means in r be best to help me achieve this? Am I right that whichever analysis I use, I use the mean latitude and longitudes for each name as below? I was reading about GPSeqClus in R which seems to be more suited to GPS data but that is way beyond me.

# Select longitude and latitude
coordinates <- mean_coordinates[, c("long", "lat")]
# Perform hierarchical clustering
hc <- hclust(dist(coordinates))

data example:

structure(list(name = c("5002", "1003", "1206", "0044", "0003"), group= c("a", "a", "a", "b", "b"), lat = c(50.406116716098, 50.4025135598379, 50.4029167297529, 50.409181401875, 50.3945484759622 ), long = c(-4.89126066209428, -4.88438821091041, -4.88681572688751, -4.84956106473437, -4.88448928953403)), row.names = c(1L, 3L, 4L, 5L, 6L), class = "data.frame")

I have lost a lot of time trying to understand all this! (complete beginner).