Ships traffic encounter scenarios generation using sampling and clustering techniques

Victor Bolbot, Christos Gkerekos, Gerasimos Theotokatos

Research output: Contribution to conferencePaperpeer-review

3 Downloads (Pure)

Abstract

The Marine Autonomous Surface Ships (MASS) constitute a novel type of systems, which require novel methods for their design and safety assurance. The collision avoidance system is considered one of the most critical systems for MASS. This study aims at developing a process for generating and selecting ship encounter scenarios to test the collision avoidance system in a virtual environment. The proposed process employs sampling techniques for generating encounter scenarios, deterministic criteria for identifying the hazardous scenarios, risk metrics estimation for the classification of the encounter situations, as well as clustering techniques for further downsizing of the scenarios number. This process is applied to a small short-shipping vessel thus demonstrating its applicability.
Original languageEnglish
Number of pages8
Publication statusPublished - 11 Jun 2021
Event1st International Conference on the Stability and Safety of Ships and Ocean Vehicles - Online
Duration: 7 Jun 202111 Jun 2021
http://www.stability-and-safety-2021.org/

Conference

Conference1st International Conference on the Stability and Safety of Ships and Ocean Vehicles
Abbreviated titleSTAB&S 2021
Period7/06/2111/06/21
Internet address

Keywords

  • marine autonomous surface ships
  • collision avoidance system
  • safety
  • testing

Fingerprint

Dive into the research topics of 'Ships traffic encounter scenarios generation using sampling and clustering techniques'. Together they form a unique fingerprint.

Cite this