Industry 4.0 corresponds in a way to the digitization of the factory. Through the use of the Internet of Things and cyber-physical systems, the intelligent factory is characterized by continuous and instantaneous communication between the various tools and workstations integrated into the production and supply chains.
The use of communicating sensors provides the production tool with a self-diagnostic capacity and thus allows its remote control, as well as its better integration into the global production system.
This project aims at the development of artificial intelligence techniques in contexts where humans and artificial intelligence together control systems and make decisions. The topics we would like to address are related to two scientific directions: sharing of control, involving in particular automated planning, multi-agent systems, and man-machine interaction; and sharing of decisions, involving in particular machine learning, knowledge extraction, and data analysis, with a privileged application in manufacturing.
The common concept that appears in this proposal is the one of explainability of artificial intelligence systems. In fact, artificial intelligence technologies, as machine learning, neural networks and deep learning, are applied in many different areas, and also in production and Industry 4.0. These techniques have proven to be powerful providing decisions and solutions in situations that have not been possible to carry out before. However, the AI-based systems are considered to be black boxes and it is difficult to follow the AI-systems reasoning, as well as investigate how they reached their conclusions.
Develop research in the area of explainable artificial intelligence systems.
The one-year project will start to analyze and conduct research within the area. The expectation is to get insight in the problem area and provide some high-level solutions. The result will be documented and presented at international conferences.