Comprehensive Assessment of Automatic Identification System (AIS) Data Application to Anti-collision Manoeuvring

Felski, Andrzej; Jaskólski, Krzysztof; Banyś, Paweł · 2015 · Crossref

DOI: 10.1017/s0373463314000897

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the limitations of radar-based collision avoidance systems, which are effective only for constant ship motion parameters and suffer from significant position errors, weather dependencies, and detection constraints. The authors propose supplementing radar with Automatic Identification System (AIS) data, which offers continuous, automatic transmission of ship motion parameters with accuracy comparable to Global Navigation Satellite Systems (GNSS). However, the study investigates the integrity and completeness of AIS information, noting that prior research often highlighted imperfections such as missing transmissions or unreliable data, particularly regarding static and voyage-related information. The methodology involves a comprehensive statistical analysis of AIS data integrity and completeness using reliability theory models. The authors define "completeness" as the degree of compatibility with ITU-R M.1371 specifications and "integrity" as the confidence in the correctness of the information. They analyzed AIS datasets from two primary regions: the Baltic Sea (east of Bornholm, between Trelleborg and Arkona) and the Gulf of Gdańsk. For the Gulf of Gdańsk, data was recorded from April 2006 to January 2012 using a receiver in Gdynia. The study focused on dynamic information, specifically Heading (HDG) and Rate of Turn (ROT), as well as Speed Over Ground (SOG) and Course Over Ground (COG). The authors applied two criteria for assessment: a "lines" criterion (counting VDM message lines) and a "ships" criterion (counting vessels transmitting incorrect data). They also utilized exponential distribution models to calculate completeness coefficients based on working and failure times of the information system. The results indicate that AIS data quality varies significantly by location and vessel speed. In harbor areas, approximately 35% of AIS messages contained unknown values for analyzed variables, whereas in sea areas, only about 8% were affected. The highest occurrence of unknown values was observed for ROT and HDG, particularly when vessel speed was less than 1 knot. In the Gulf of Gdańsk, analysis showed that 16% of messages lacked HDG information and 17% lacked ROT information under the "lines" criterion. Under the "ships" criterion, 20.5% of vessels transmitted messages lacking HDG, and 22% sent incomplete ROT messages. However, the study noted a significant improvement in AIS information quality over time, with incompleteness coefficients dropping below 0.10 for HDG and ROT data in the 2010–2012 period. The authors conclude that while AIS data has historically suffered from incompleteness, particularly in harbor areas and at low speeds, it can deliver useful supplementary information for collision avoidance. The study suggests that dynamic information derived from GNSS receivers, such as Position, COG, and SOG, is the most reliable for determining Closest Point of Arrival (CPA) and Time to CPA. The findings support the integration of AIS with radar systems to enhance navigational safety, provided that users account for the specific integrity limitations of dynamic parameters like HDG and ROT.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• naturalistic crash near crash