We recently learned that the project "RICO- Robust Intelligent Control", will receive funding from the Norwegian Research Council (2022-2025). The ambition of the project is to develop new methods for improved robustness of intelligent control systems. 

Intelligent control systems are increasingly being deployed in safety-and business-critical applications. Such systems perform the tasks of traditional control systems; they manage and command devices and other systems. In addition, intelligent control systems apply artificial intelligence as a means to achieve adaption, learning and planning under uncertainty. 

When dealing with intelligent control systems, being able to trust the system's intelligent characteristics is crucial. We often refer to this as the AI system's trustworthiness. Trustworthy AI is a wide term, which according to the EU refers to AI that is lawful, ethical and robust. 

In a new project, researchers from NTNU and SINTEF will pay special attention to the robustness of intelligent control systems, as a means to increase trust in the systems. 

New methods for robust intelligent control

In this setting, robustness refers to the ability to maintain performance and stability in control systems that are subject to uncertainty, disturbances and modelling errors. Although intelligent controllers have been shown to outperform traditional controllers in several domains, the usage of intelligent controllers in industry is still limited due to a lack of robustness in the underlying methodology.

There is thus a gap between the research on very advanced methods tested in a controllable and predictable environment and the industrial setting where the environment is neither completely known nor controllable and where information can be scarce, uncertain and of low quality. 

This gap can be closed by developing novel methods for robust intelligent control and forecasting to be tested in both controlled environments and in industrial systems. This is exactly the aim of the new project "RICO- Robust Intelligent Control", funded by the Norwegian Research Council, which will start up in 2022. 

Generic methods and technology- with real implications for industry 

The results, methods and technologies to be developed in the RICO project will be of a general nature. However, in order to evaluate the outputs of the project, industrial partners will have to take part. The project will therefore have practical implications in different domains. In addition to NAIL research partners, NTNU and SINTEF, TrønderEnergi is an industrial partner, together with Solution Seeker and Ohmia Charging. 

The use-cases selected from the industrial partners cover various domains; renewable energy production, process industry, smart buildings and transportation. Optimizing and controlling energy production and consumption in these domains will lead to improved utilization of energy. This will contribute to making energy more affordable, reducing the carbon footprint and enabling the transition to electric vehicles.