Using source data to aid and build variational state-space autoencoders with sparse target data for process monitoring

Article Automation & Control Systems

Body Pose Prediction Based on Motion Sensor Data and Recurrent Neural Network

Marcin Wozniak et al.

Summary: Mixed reality environments offer better opportunities for providing constant assistance to those in need. Utilizing artificial intelligence models and recurrent neural networks, the system can detect symptoms of unusual poses with high accuracy of 99.89% using minimal body sensors, helping to protect life in dangerous conditions. This system can operate without the need for special infrastructure or development, providing support for workers and elderly individuals in various environments.

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Supervised and semi-supervised probabilistic learning with deep neural networks for concurrent process-quality monitoring

Kai Wang et al.

Summary: Concurrent process-quality monitoring helps discover quality-relevant process anomalies and quality irrelevant process anomalies, especially in chemical plants. The proposed method based on a probabilistic generative deep learning model effectively outperforms other comparative methods in concurrent process-quality monitoring.

NEURAL NETWORKS (2021)

添加到收藏夹

Article Automation & Control Systems

Recurrent Neural Network Model for IoT and Networking Malware Threat Detection

Marcin Wozniak et al.

Summary: The security of networking in cyber-physical systems is crucial, and this article proposes a solution based on a deep learning model for analyzing network traffic. The model efficiently evaluates information and makes security decisions, achieving over 99% accuracy even with a reduced number of features evaluated.

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS (2021)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications