Thin a track by along-track distance

Select a subset of fixes spaced by cumulative along-track distance, returning the move2 object whole with a logical column thin_selected marking retained fixes. Two methods are offered:

Usage

mt_thin_distance(
  x,
  distance,
  tolerance = NULL,
  criterion = c("closest", "first"),
  method = c("step", "interval"),
  remove = FALSE
)

Arguments

x

A move2 object.

distance

The target along-track distance step, in metres. Accepts a numeric scalar or a units object convertible to metres (e.g. units::set_units(0.5, "km")).

tolerance

Half-width of the acceptance window around distance, in metres. Used only by method = "interval"; ignored (with a warning if set) by method = "step". Defaults to distance / 10.

criterion

Tie-breaking rule for method = "interval" when multiple chains of equal length exist: "closest" (minimum total deviation from the target, default) or "first" (earliest-starting chain). Ignored by method = "step".

method

Thinning rule: "step" (default, greedy minimum-spacing) or "interval" (tolerance-constrained longest chain).

remove

Logical. If TRUE, return only the retained fixes. Default FALSE – the full object is returned with a logical thin_selected column.

Value

A move2 object. With remove = FALSE, the input is returned unchanged except for a new logical column thin_selected. With remove = TRUE, only rows with thin_selected == TRUE are returned.

Details

• method = "step" (default) walks the track from the first fix and retains a fix each time the cumulative distance since the last retained fix reaches distance. Every retained pair is therefore at least distance of travel apart – the intuitive "give me one fix per distance of travel" behaviour.

• method = "interval" is the distance-space twin of mt_thin_time(): it finds the largest subset such that every successive retained pair is within [distance - tolerance, distance + tolerance] of along-track travel. This is the faithful move2 analogue of move::thinDistanceAlongTrack(), which is itself the distance counterpart of move::thinTrackTime().

Per-segment distance comes from move2::mt_distance(units = "m"), so distance (and tolerance) are always in metres regardless of the input CRS's linear unit. Geographic (lon/lat) data are thinned against great-circle distance and projected data against planar distance, so the two can differ slightly at the same metre threshold; project to an equal-area metric CRS first if you need the result to be identical across coordinate systems. Multi-track input is thinned independently per individual.

For method = "step", the first fix of each track is always retained, and – because the threshold is crossed mid-segment and we do not interpolate – the realised distance between successive retained fixes can exceed distance (it never falls below it). Use move2::mt_interpolate() first if you need exact-distance spacing.

For method = "interval", the track is pre-split at any single step longer than distance + tolerance (thinning never bridges such a step), and the longest tolerance-admissible chain is selected by the same dynamic-programming sweep used by mt_thin_time(). The first fix is not guaranteed to be retained, and no retained pair is closer than distance - tolerance.

See also

mt_thin_time() for time-based thinning (same DP machinery); move2::mt_interpolate() to first regularise the track in time or along a target line; move2::mt_distance() for the underlying distance.

Examples

# \donttest{
library(move2)
fishers <- mt_read(mt_example())
fishers <- fishers[!sf::st_is_empty(fishers), ]
leroy <- fishers[mt_track_id(fishers) == "M4", ][seq_len(500), ]

## one fix per ~300 m of travel (minimum spacing guaranteed)
out <- mt_thin_distance(leroy, distance = 300)
table(out$thin_selected)
#> 
#> FALSE  TRUE 
#>   443    57 

## tolerance-constrained: retained pairs ~300 m +/- 30 m apart
out2 <- mt_thin_distance(leroy, distance = 300, tolerance = 30,
                         method = "interval")
table(out2$thin_selected)
#> 
#> FALSE  TRUE 
#>   448    52 
# }