Suchergebnisse

This article shows a way to store the data of many S-101 charts into a single Postgres database. The data model of the database with all tables is shown and explained. The concatenation of the indices from the different database tables is explained. This concatenation allows for a faster search of points/curves with certain properties. This fulfills one of the basic requirements for the purpose of navigating an autonomous vessel – that several charts can be interpreted simultaneously by a machine. Mechanisms for up-dating the database with new charts not yet present in the database are shown. Also the mechanisms for updating the charts already present in the database are explained. System limitations are briefly presented to show that in practical use there are in fact none. Memory requirements for such a type of chart storage in the database is compared with memory requirements for ISO8211 files normally used for storage of S-101 charts. With small examples it is finally shown how the stored chart information can be searched specifically.

The state of prices and the development of a given economy depends largely on the gas and oil industry. Every day the gas and oil industry is growing and there is a growing need for platforms. In order for the platforms to pump gas and oil from the seabed unhindered, it is necessary to supply the platform which is done by supply vessels. In this paper, with the help of Petri nets and the tool "Yasper", the dynamic process of supply and operation of supply vessels is simulated. Simulation is depending on the wind and wave conditions and the type of work that the supply vessel performs in cooperation with the platform. Input data are: Sea state, wind state, fuel consumption, type of vessel operation, distance from the mainland and the number of daily departures of the vessels from the port. The paper presents a model with discrete events in the process of supplying the platform, starting from the port and going out to sea until the work around the platform is completed and returning to the port. The report can provide information on the usability of the vessels and the approximate cost of the supply vessels on a voyage from port to the platform and back.

Automatic Identification Systems (AIS) and Automatic Radar Plotting Aids (ARPA) are commonly used to detect targets for collision avoidance. However, AIS cannot detect targets without AIS transmitters and ARPA has limitations due to blind sector and small targets may not be detected. Advances in computer performance and video-based detection generated much interest in developing intelligent video surveillance systems to achieve autonomous navigation. To develop a reliable collision avoidance system, we propose the use of a visual camera for real-time object detection and target tracking. Moreover, the system should follow the International Regulations for Preventing Collisions at Sea (COLREGs) to avoid catastrophic accidents. In this paper only a part of the system is presented. For real-time object detection, the You Only Look Once (YOLO) ver. 3 convolutional neural network is used, and the target tracking filter based on a Kalman filter with built-in estimated relative position and velocity.