Indonesia requires a maritime surveillance system that is capable for monitoring its wide waters territory. Indonesian National Institute of Aeronautics and Space (LAPAN) with LAPAN-A2 and LAPAN-A3 satellites which have Automatic Identification System (AIS) receiver as payloads, make a significant contribution to maritime surveillance in Indonesian territory. The AIS data receives from LAPAN Satellite is raw form data, such as the ship identity (static message) and navigation status (dynamic data), which need to be processed further to maximize the usage and able to provide certain information. This study focuses on implementation of unsupervised learning For AIS Data of LAPAN Satellites using Density-Based Spatial Clustering of Applications with Noise (DBSCAN) with additional parameter to extract information like characteristics of the ship's shipping lane, ship behaviors and find hidden patterns in AIS data of LAPAN Satellite. The results of the analysis show that by adding two new parameters (COG and SOG) to DBSCAN, the results of the clustering performed was greatly improved. The results of optimized DBSCAN provide a dataset that can be used to discover vessel behavior and the characteristics of shipping lanes in an area.
Implementation of Unsupervised Learning Based On AIS Data of LAPAN Satellites
03.11.2021
1501641 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Preliminary study of the potential utilization of imagery from LAPAN-A4 satellites
American Institute of Physics | 2021
|