Collision risk-informed weather routing for sailboats
•A route planning for sailboats with collision avoidance is presented.•Collision risk is avoided based on a configurable domain of a motor-driven ship.•Ship domain is mapped to a graph, where some edges are temporarily unavailable.•Wave resistance is taken into account when assessing sailboat's...
Gespeichert in:
Veröffentlicht in: | Reliability engineering & system safety 2023-04, Vol.232, p.109015, Article 109015 |
---|---|
Hauptverfasser: | , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | •A route planning for sailboats with collision avoidance is presented.•Collision risk is avoided based on a configurable domain of a motor-driven ship.•Ship domain is mapped to a graph, where some edges are temporarily unavailable.•Wave resistance is taken into account when assessing sailboat's speed.•The method is implemented as a configurable software fast enough for real time use.
Selected COLREG rules, good seamanship and sheer common sense indicate that it is in a sailboat's interest to follow collision-free routes without relying on large power-driven ships to give way. Until now, however, no method has integrated a sailboat's weather routing with collision risk monitoring and collision avoidance. Therefore, a new deterministic approach to combine the above features within one method is introduced here. The proposed method is based on Dijkstra's algorithm, where edges may be temporarily removed due to the presence of other ships. This paper presents a design of the main weather routing algorithm and the collision risk monitoring part, which applies an elliptic domain generated automatically around the target and dependant on the target's length. The method has been implemented and tested in a series of computer simulations. The results are provided and discussed here. They confirm the method's effectiveness in terms of determining collision-risk-free routes, as well as its acceptable computational time. They also show how the latter can be shortened at the cost of obtaining suboptimal routes. Finally, they emphasize the importance of considering successive weather forecasts, risk monitoring and route updates. |
---|---|
ISSN: | 0951-8320 1879-0836 |
DOI: | 10.1016/j.ress.2022.109015 |