Refereed article in conference proceedings (A4)
Ship Movement Prediction Using k-NN Method
List of Authors: Petra Virjonen, Paavo Nevalainen, Tapio Pahikkala, Jukka Heikkonen
Editors: Randall Bilof
Conference name: Baltic Geodetic Congress, Geomatics
Publication year: 2018
Book title *: 2018 Baltic Geodetic Congress (BGC Geomatics 2018). 21-23 June 2018, Olsztyn, Poland. Proceedings
Start page: 304
End page: 309
ISBN: 978-1-5386-4898-8
DOI: http://dx.doi.org/10.1109/BGC-Geomatics.2018.00064
Trajectories of ships travelling in the Gulf of
Finland were predicted using the k-Nearest Neighbours (k-NNs)
method. Automatic Identification System (AIS) data gathered via
open interface of the Finnish Transport Agency were used. The
results will be exploited in a route optimization task for an
emission control boat. The task requires prediction several hours
ahead with reasonable accuracy.
The idea is to compare the trajectories of a new ship and
historical ships within a comparison area. The future behaviour of
the new ship was estimated with the k-nearest neighbours. The
performance of the method as well as the hyper parameters
(nearest neighbours, k, and a weighting parameter Į) of the
proposed model were optimized using nested leave-one-out crossvalidation
approach.
The method enables the prediction within minutes’ accuracy
in time and less than 2 km in location several hours ahead, which
is more than satisfactory for the route optimization purposes.