Using machine learning for fault detection in lighthouse light sensors
Michael Kampouridis, Nikolaos Vastardis, George Rayment
TL;DR
The study tackles fault detection for lighthouse photoresistor sensors in the absence of labeled failure data by simulating gradual sensor faults and evaluating four classifiers on climate-augmented features. The multi-layer perceptron (MLP) emerges as the best performer, capable of identifying timing drifts within 10–15 minutes and serving as an early warning tool. Baseline classification across seven lighthouses yields 80–85% accuracy (Godrevy ~96–97%), while drift experiments reveal a consistent decay in accuracy and F1 scores as drift increases, demonstrating the method’s practical potential for proactive maintenance planning. The work supports automated, data-driven prioritization of maintenance visits to improve navigational safety and reduce operational costs, with future work exploring a global model that aggregates data across lighthouses.
Abstract
Lighthouses play a crucial role in ensuring maritime safety by signaling hazardous areas such as dangerous coastlines, shoals, reefs, and rocks, along with aiding harbor entries and aerial navigation. This is achieved through the use of photoresistor sensors that activate or deactivate based on the time of day. However, a significant issue is the potential malfunction of these sensors, leading to the gradual misalignment of the light's operational timing. This paper introduces an innovative machine learning-based approach for automatically detecting such malfunctions. We evaluate four distinct algorithms: decision trees, random forest, extreme gradient boosting, and multi-layer perceptron. Our findings indicate that the multi-layer perceptron is the most effective, capable of detecting timing discrepancies as small as 10-15 minutes. This accuracy makes it a highly efficient tool for automating the detection of faults in lighthouse light sensors.