Funded Research

The MoReTrust project focuses on developing methods and tools to enhance the resilience of modern software systems against security threats. As smart software systems become more complex and interconnected, ensuring their trustworthiness and protecting them from security threats is increasingly challenging. The project emphasizes the importance of addressing not only known threats but also potential vulnerabilities that may arise during the system's operation.

MoReTrust explores the use of digital twins as system doubles to provide virtualization and create a protective layer against security breaches. The project combines model-driven engineering (MDE), control theory, and digital twins to create a comprehensive framework for trustworthy smart software systems. This framework supports the identification of uncertainties and threats, the design of self-adaptive mechanisms (such as adaptation, healing, protection, and reconfiguration), and the runtime detection and response to these threats.

Duration: 2024-2027

This project will develop algorithms and tools to provide a generic federated learning (FL) framework to deploy trustworthy and energy-efficient machine learning (ML) applications onto heterogeneous edge devices. It uses a cross- layer approach to utilize the full potential of the FL processing flow from communication scheduling to hardware deployment. To this aim, we will (i) develop adaptive scheduling mechanisms that guarantee the system predictability; (ii) design scalable and energy-efficient heterogeneous machine learning (HML) models customized for each FL node individually with guaranteeing user requirements; (iii) design trustworthy FLs that are resistant to different types of cyber-attacks as well as noisy environments; and (iv) facilitate fast deployment and effective maintenance of HML models.

Duration: 2024-2027

The TRIM (Trustworthy Interaction of Multi-agent Systems) project aims to enhance the security of Cyber-Physical Systems (CPS) amid increasing digital interconnectivity. By treating distributed CPSs as Multi-Agent Systems (MAS), TRIM develops innovative control-based approaches to counter sophisticated cyber threats like stealth attacks. The project integrates advanced mathematical theories and robust optimization to create scalable, explainable, and efficient security solutions. These solutions include mechanisms for preventing, detecting, and responding to attacks at both single-agent and multi- multi-agent levels, ensuring secure interactions even in the presence of malicious entities. TRIM's outcomes will significantly advance CPS security, contributing to more resilient infrastructure systems.

Duration: 2024-2027

The NeuroTrust project aims to improve motor recovery for chronic stroke patients with severe motor impairments using Brain-Computer Interface (BCI) technology. The project has three main goals: identifying neural mechanisms for motor recovery, optimizing neural learning with real- time Al, and developing personalized BCI strategies for functional cortical reorganization. NeuroTrust collaborates with clinical researchers and aims to create a trusted neurorehabilitation system that can be integrated into existing stroke rehabilitation programs. The project builds on promising results from a 2019 pilot study and aims to benefit stroke patients by developing a clinical prototype for physical rehabilitation.

Link: NEUROTRUST

Duration: 2024-2027

Cardiovascular diseases are a major global health challenge that requires early detection and continuous monitoring for effective treatment. The primary objective in this project is to create an Al-based reliable smart system for real-time monitoring of cardiovascular health. This system will allow for accurate, non-invasive measurements and predictions of physiological parameters such as blood pressure and arterial stiffness, based on photoplethysmography (PPG) signals. The project involves developing new pre-processing techniques and feature extraction algorithms for PPG signals, designing an Al-driven real-time monitoring system, implementing a prototype of the system, and integrating Al algorithms with PPG signal analysis to enhance the accuracy and reliability of cardiovascular health assessments.

Duration: 2024-2027

The DOSTY project aims to develop a three-layered dependability assurance framework for Trusted Intelligent System of Systems (TriSoS), focusing on safety and security. This framework includes a safety controller, system dependency modeling, fault propagation analysis, and an anomaly detection algorithm to ensure robust protection against cyber threats. The framework will be prototyped and verified through simulations, demonstrating the feasibility of integrating multiple intelligent systems while prioritizing safety and security.

Duration: 2025-2028

The INSID project aims to create a scalable framework that combines formal methods and machine learning (ML) to ensure the safety and efficiency of autonomous driving functions. ML algorithms help by guiding the exploration of complex driving scenarios and generating realistic traffic situations, including rare events. This approach ensures a consistent environment model and enables adaptive behaviors for safety. The framework aims to detect errors and unexpected behaviors early in the development of autonomous vehicles. The project will develop a proof-of- concept toolchain and demonstrate it with industrial use cases from Volvo Cars.

Duration: 2025-2028

The project aims to systematically investigate the effectiveness of existing trustworthiness measures on smart production systems. It involves designing and developing a trusted and reconfigurable production system by integrating soft programmable logic controllers (PLCs), which can be used alongside conventional PLCs to ensure high availability and trust.

Additionally, the project seeks to connect the developed soft PLCs and devices via 5G or WiFi6 to leverage new technologies with high reliability.

Finally, the project aims to devise and implement a methodology to evaluate trustworthiness when adding new devices and connectivity solutions. Part of this tation is being implementation is being carried out on the FESTO line at MITC by integrating a soft PLC, which will be further enhanced by developing integrated solutions and connectivities.

Duration: 2025-2028