This paper addresses anomaly detection and monitoring for swarm drone flights. While the current practice of swarm flight typically relies on the operator’s naked eyes to monitor health of the multiple vehicles, this work proposes a machine learning-based framework to enable detection of abnormal behavior of a large number of flying drones on the fly. The method works in two steps: a sequence of two unsupervised learning procedures reduces the dimensionality of the real flight test data and labels them as normal and abnormal cases; then, a deep neural network classifier with one-dimensional convolution layers followed by fully connected multi-layer perceptron extracts the associated features and distinguishes the anomaly from normal conditions. The proposed anomaly detection scheme is validated on the real flight test data, highlighting its capability of online implementation.

A spacecraft attitude control system provides mechanical and electrical control to achieve the required functions under various mission scenarios. Although generally designed to be highly reliable, mission failure can occur if anomalies occur and the attitude control system fails to properly orient and stabilize the spacecraft. Because accessing spacecraft to directly repair such problems is usually infeasible, developing a continuous condition monitoring model is necessary to detect anomalies and respond accordingly. In this study, a method for detecting anomalies and characterizing failures for spacecraft attitude control systems is proposed. Herein, features are extracted from multidimensional time-series data of a simulation of the attitude control system. Then, the artificial neural network learning algorithms based on two types of generation models are applied. A Bayesian optimization algorithm with a Gaussian process is used to optimize the hyperparameters for the neural network to improve the performance. The performance is evaluated based on the reconstruction error through the algorithm using the newly generated data not used for learning as input data. Results show that the detection performance depends on the operating characteristics of each submode in the operation scenarios and type of generation model. The diagnostic results are monitored to detect anomalies in operation modes and scenarios.

As the workforce shrinks, the demand for automatic, labor-saving, anomaly detection technology that can perform maintenance on advanced equipment such as vehicles has been increasing. In a vehicular environment, noise in the cabin, which directly affects users, is considered an important factor in lowering the emotional satisfaction of the driver and/or passengers in the vehicles. In this study, we provide an efficient method that can collect acoustic data, measured using a large number of microphones, in order to detect abnormal operations inside the machine via deep learning in a quick and highly accurate manner. Unlike most current approaches based on Long Short-Term Memory (LSTM) or autoencoders, we propose an anomaly detection (AD) algorithm that can overcome the limitations of noisy measurement and detection system anomalies via noise signals measured inside the mechanical system. These features are utilized to train a variety of anomaly detection models for demonstration in noisy environments with five different errors in machine operation, achieving an accuracy of approximately 90% or more.

Our study focuses on leveraging a deep neural network approach coupled with geostationary satellite data to enhance short-term three-hour-ahead predictions of solar radiation over Northeast Asia. A crucial aspect of our study was improving the smoothness of the predicted image patterns, especially in cases of long-term forecasts, based on the application of the HRNet model. The proposed method effectively depicted 2D images of the predicted maps for potential solar radiation because it maintains high-resolution representations throughout its layers. Our short-term prediction maps for potential solar radiation showed good agreement with reference images from the physical model, and complex cloud movements were well predicted using the HRNet model. Reliable accuracy was obtained between the ground pyranometer measurements and our predicted values for the potential three-hour-ahead predictions (RMSE = 98.570 Wm ⁻² , nRMSE=23.79 %, MBE = −0.578 Wm ⁻² , and R ² = 0.829). Furthermore, the HRNet model demonstrated a notably low average blurriness score of 0.00019 and high average peak signal-to-noise ratio of 7.646, demonstrating its improved ability to deliver sharper and less-noisy predictions than selected deep learning models.