Today most ship rotation angle (steering control during movement) increase or decrease is done using an operator on deck or the auxiliary system in the ships engine room. Formal regulations suggest using manual inspection of the ship rotation and the work effectiveness of the engine during manoeuvring in ports and in the open sea regions. The accuracy of this procedure is very low and depends on the personnel of the deck. Therefore, automation and computer control systems are constantly required to assist the human eye. This problem becomes clearly visible when dealing with full ship autonomy in the open sea in the short-sea shipping regions. The trend of maritime technology development will only increase in the area of human interaction decrease with the physical operations and the shipping procedures, which will lead to the future full ship autonomy in the open sea regions around the globe. With the growing automation technologies, predictive control can prove to be a better approach than the traditionally applied visual inspection policy and linear control models. Ship full autonomy is also linked to the ship's machinery regular repair and maintenance that has to be carried out for delivering satisfactory performance and minimizing downtime during transportation operations. In this paper, current stages of development of the intelligent transportation system concept are discussed for the ship autonomy in manoeuvring control and a robust ships' systems integration and communication system concept is presented for several normal and abnormal situations: high-traffic, potentially dangerous situations or port approaching or ship maintenance, with the capability to solve problems with the limited human interface and with a remote control possibility. Then, simplified ship steering motor system for the main pump is analysed for rotation control using control voltage from the converters. Retrieved data from a small experimental control motor is used for the predictive control approach using two different methods: a neural network trained with Basic Levenberg-Marquardt Method and a Linear Model.
Safe navigation at sea is more important than ever. Cargo is usually transported by vessel because it makes economic sense. However, marine accidents can cause huge losses of people, cargo, and the vessel itself, as well as irreversible ecological disasters. These are the reasons to strive for safe vessel navigation. The navigator shall ensure safe vessel navigation. He must plan every maneuver and act safely. At the same time, he must evaluate and predict the actions of other vessels in dense maritime traffic. This is a complicated process and requires constant human concentration. It is a very tiring and long-lasting duty. Therefore, human error is the main reason of collisions between vessels. In this paper, different reinforcement learning strategies have been explored in order to find the most appropriate one for the real-life problem of ensuring safe maneuvring in maritime traffic. An experiment using different algorithms was conducted to discover a suitable method for autonomous vessel navigation. The experiments indicate that the most effective algorithm (Deep SARSA) allows reaching 92.08% accuracy. The efficiency of the proposed model is demonstrated through a real-life collision between two vessels and how it could have been avoided.
During the last years, marine traffic dramatically increases. Marine traffic safety highly depends on the mariner’s decisions and particular situations. The watch officer must continuously observe the marine traffic for anomalies because the anomaly detection is crucial to predict dangerous situations and to make a decision in time for safe marine navigation. In this paper, we present marine traffic anomaly detection by the combination of the DBSCAN clustering algorithm (Density- Based Spatial Clustering of Applications with Noise) with k-nearest neighbors analysis among the clusters and particular vessels. The clustering algorithm is applied to the historic marine traffic data – a set of vessel turn points. In our experiments, the total number of turn points was about 3 million, and about 160 megabytes of computer store was used. A formal numerical criterion to com-pare anomaly with normal traffic flow case has been proposed. It gives us a possibility to detect the vessels outside the typical traffic pattern. The proposed meth-od ensures the right decisions in different oceanic scale or hydro meteorology conditions in the detection of anomaly situation of the vessel.
Machine learning is compelling in solving various applied problems. Nevertheless, machine learning methods lack the contextual reasoning capabilities and cannot be fitted to utilize additional information about circumstances, environments, backgrounds, etc. Such information provides essential knowledge about possible reasons for particular actions. This knowledge could not be processed directly by either machine learning methods. This paper presents the context-aware machine learning approach for actor behavior contextual reasoning analysis and context-based prediction for threat assessment. Moreover, the proposed approach uses context-aware prediction to tackle the interaction between actors. An idea of the technique lies in the cooperative use of two classification methods when one way predicts an actor’s behavior. The second method discloses such predicted action (behavior) that is non-typical or unusual. Such integration of two-method allows the actor to make the self-awareness threat assessment based on relations between different actors where some multidimensional numerical data define the connections. This approach predicts the possible further situation and makes its threat assessment without any waiting for future actions. The suggested approach is based on the Decision Tree and Support Vector Method algorithm. Due to the complexity of context, marine traffic data was chosen to demonstrate the proposed approach capability. This technique could deal with the end-to-end approach for safe vessel navigation in maritime traffic with considerable ship congestion.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.