Decision Support System for Diagnosing Underwater Electrical Cables

Abstract

The object of this study is the process of generating an appropriate response by an intelligent agent when detecting and tracking an underwater electrical cable using a decision support system. The goal of the work is to develop the architecture of a decision support system for diagnosing underwater electrical cables and the main algorithms for its operation. To achieve the research goal, intelligent agents with a complex IPK architecture, which includes information, preferences (rules), and knowledge, were used. The combination of the structure of intelligent agents and the advantages of hierarchical knowledge bases allowed for the natural language representation of knowledge. The functional core of the system consists of four main agents, one of which facilitates the interaction between the user and the surrounding environment. To address the uncertainty in the cable's position on the seabed, the capabilities of fuzzy sets, describing its feature space with membership functions, were employed. The most significant results involve planning for the detection and tracking of underwater electrical cables, taking into account past decision-making experiences in similar cases and adapting them to the current situation through case-based reasoning. The importance of the obtained results lies in providing the decision-maker with a possible option for deploying an underwater vehicle based on accepted logic and rules for detecting underwater electrical cables. Further research is focused on implementing automatic target selection for intervention and developing a method for automatic modification of reasoning methods, preference rule bases, and operational knowledge.

Keywords

• electrical cable
• diagnostics
• autonomous underwater vehicle
• navigation
• motion control
• fuzzy sets
• intelligent agent
• decision support system.