Projects:

• WP5: Design of Security Functionality for V2X Networks. IAF PP on Connected Smart Mobility (COSMO), 2019-2022.

• Discrete Event Based Cyber Security Analysis and Attack Resilient Controller Synthesis for Cyber Physical Systems, 2018 - 2021, completed.
  
  Abstract: The Industry 4.0 and Smart Cities initiatives around world heavily rely on real-time data collection and decision-making through large-scale ICT infrastructures, which makes cyber security one of the core features required by any futuristic industrial system. In parallel with efforts from many other communities such as the control community, the embedded system community and the computer science community, in this project we address the cyber security issues from a discrete-event perspective, aiming to (1) understand what could be “intelligent” attacks; (2) describe attack mechanisms in a formal discrete-event modelling framework; and (3) design controllers capable of being resilient to those attacks. Some specific types of attacks under consideration include actuator attacks, sensor attacks, communication delay attacks and their combinations on both sensor channels and actuation channels. Modelling, analysis and attack/supervisor synthesis strategies have been developed.

• Secure and Privacy Preserving Multi-agent Cooperation, 2017 - 2019, completed.
  
  Abstract: Multi-agent systems, such as wireless sensor networks, multi-vehicles, data centers, smart grids, and multi-robots, as a simple model of cyber-physical system, have been extensively studied over the past decade. What makes the cooperative behaviors interesting is that we always want to seek benefits from limited resources we have and limited information exchange, which are also the motivation for us to study new cooperative control technologies, or called distributed control algorithms, to explore resources that may not be fully utilized under the traditional control framework. Within the framework of multi-agent systems, in general, considering the number of the specific leaders in realistic systems, the cooperative behaviors can be categorized into synchronization problems (without any leader), consensus problems (with one leader), and containment control problems (with multiple leaders). Accordingly, consensus/synchronization, as one of important cooperative behaviors, has attracted a great number of research results for the agents with first-order, second-order, higher-order, linear dynamics, and nonlinear systems dynamics. However, even though a variety of algorithms are proposed to accomplish specific cooperation tasks in different applications, the security and privacy aspects of these algorithms have not received due attention until most recently. Secure cooperation implies that the existence of parts of agents who do not follow nominal cooperation rule can be acted frequently and the influence of these agents on the cooperation task can be eliminated or compensated. Privacy preserving cooperation means that each agent does not disclose any information more than necessary to achieve the cooperation task or disguise as different dynamics to block detection. The contribution of this project is to analyze and design secure and privacy preserving multi-agent cooperation algorithm in fault and heterogenic dynamics situations. To this end, we have studied to: I) secure preserving multi-agent cooperation subject to actuator faults; I I) secure preserving multi-agent cooperation from against frequency communications; I I I) privacy preserving multi-agent cooperation under heterogeneous models; and V) privacy preserving multi-agent cooperation with an unknown leader.