donut {donut}R Documentation

donut: Nearest Neighbour Search with Variables on a Torus


Finds the k nearest neighbours in a dataset of specified points, adding the option to wrap certain variables on a torus. The user chooses the algorithm to use to find the nearest neighbours.


The function nnt performs the nearest neighbour search. There is also a rudimentary plot method: plot.nnt.

The default algorithm is that provided by the function nn2 in the RANN-package. Another possibility is the knn function in the nabor-package.

See vignette("donut-vignette", package = "donut") for an overview of the package.


Arya, S., Mount, D., Kemp, S. E. and Jefferis, G. (2019) RANN: Fast Nearest Neighbour Search (Wraps ANN Library) Using L2 Metric. R package version 2.6.1.

Elseberg J., Magnenat S., Siegwart R., Nuchter, A. (2012) Comparison of nearest-neighbor-search strategies and implementations for efficient shape registration. Journal of Software Engineering for Robotics (JOSER), 3(1), 2-12

See Also

nnt for nearest neighbour with some variables wrapped on a torus.

plot.nnt plot method for objects returned from nnt (1 and 2 dimensional data only).

[Package donut version 1.0.2 Index]