Intelligent Algorithms for Fault Detection and Recovery in Autonomous Submersibles

Abstract

Fault detection and recovery are critical for the safe and reliable operation of Autonomous Submersibles (ASs). This paper presents a novel framework utilizing intelligent algorithms for real-time fault detection, diagnosis, and recovery strategies in ASs. We explore machine learning techniques, including anomaly detection and classification algorithms, to identify faults in various subsystems and environmental conditions. The proposed framework integrates redundancy and fail-safe mechanisms to ensure mission continuity in the event of a fault. Experimental results demonstrate the effectiveness of these intelligent algorithms in enhancing the robustness and reliability of AS operations, paving the way for more autonomous underwater missions. This research contributes to advancing the operational capabilities of ASs, particularly in complex underwater environments where traditional monitoring approaches may be insufficient.