Page 95 - Demo

Automatic Identification System Data Fusion83Marine Research of the National Academy of Marine ResearchDOI: 10.29677/MR.202406_4(1).0006 In cooperation with the MOTC to promote the Smart Navigation and Safety Service Development Plan, since 2018, the Maritime and Port Bureau has been leading the integration of AIS. Based on the national lighthouse locations, 19 beacon stations and 14 base stations have been established. These stations provide services such as electronic navigation, early warning, and monitoring to enhance vessel navigation safety, reduce maritime incidents, strengthen maritime rescue operations, monitor traffic flow in Taiwanese waters, and continuously assess navigation risks for the planning and adjustment of beacon and navigational systems. Additionally, the system has integrated the existing 16 AIS receiving stations to serve as signal reinforcement and backup. According to the ITU-R M.1371-5 recommendation (Murray and Perera, 2021), there are 27 defined data types for AIS data, and each data type has its own defined data field format. This study utilizes various AIS static and dynamic data, along with their definitions and data formats, as shown in Table 1 belo The AIS operates in three modes (Wijaya and Zhang, 2013; Perera et al., 2015; Wei et al., 2020). The first mode is automatic continuous transmission of vessel information to external entities. The second mode involves vessels providing monitoring capabilities to the competent authorities within their jurisdiction. The third mode involves mutual data exchange and response between vessels and the competent authorities. The purposes of the AIS are to provide vessel identification, aid in tracking vessel targets, simplify vessel information exchange, and prevent vessel collisions. AIS messages provide two types of data: static information and dynamicinformation (Goerlandt and Kujala, 2014; Xiao et al., 2015; Xin et al., 2019). Static information includes vessel details such as vessel name, IMO number, MMSI, vessel type, length, width, and more. Dynamic information includes data like longitude, latitude, heading, speed, and other relevant parameters. The AIS enables accurate and timely reporting of vessel information. AIS electronic signals are transmitted through the air to shore-based receiving stations. These signals are then transferred to the processing center via a wired network for decoding and computation (Dogancay et al., 2021; Qu et al., 2019; Zaman et al., 2015). The processing center utilizes the AIS messages to build a database, which incorporates static and dynamic vessel data, weather data, electronic charts, and other relevant information (Rawson & Brito, 2021). This database is analyzed and processed to provide the latest information services to various government and search and rescue agencies (Mao et al., 2018; Rong et al., 2015; Wang et al., 2013). As mentioned in the preceding section regarding the current development of AIS technology and the goals of this study, our focus is on establishing a quality management mechanismfor the AIS database. The aim is to establish an operational and standardized AIS data quality control mechanismthat effectively identifies and eliminates abnormal or inconsistent AIS data within the system. In related researchon AIS data cleaning and management, a spatial clustering method (SPTCLUST-II) is introduced. This study clusters a substantial amount of ship trajectory data in a short time by extracting historical data from ship AIS, enabling an understanding of maritime vessel navigation behavior. This, in turn, may contribute significantly to the benefits of ship traffic management in intelligent transportation systems. The study also outlines a method for clustering vessel trajectory segments and extracting maritime traffic routes, comprising four main parts. The first part involves the collection and cleaning of AIS data. Initially, the study includes the removal of duplicate data in the collected raw AIS data. Through the use of GIS applications, the original AIS data was verified, encompassing checking its spatial distribution, comparing data discrepancies within the same vessel AIS entry, and removing observations outside the target area and land-based observations.