Home // Research // FSTM // DCS // Research Pro... // Decentralized global decision-making over dynamic networks of proactive engines

Decentralized global decision-making over dynamic networks of proactive engines

Funding: University of Luxembourg
Start Date: Nov. 1, 2019
End Date: Nov. 1, 2022


Proactive Computing is a recent research field, which aims at the development of new IT systems and software applications that work in a more autonomic way for the user's interests. Based on predefined scenarios, the system decides alone about its actions for reacting in a swift and best appropriate way to the changes in its environment, without the command of human beings. Implementing such complex systems into large and/or complex real-world environments often requires one to connect several proactive engines over a dynamic network, for multiple reasons such as geographic proximity of the engines with sensors or actuators, specific computing capacities in engines, redundancy of engines for safer robustness, etc. Each proactive engine taking its decisions locally and acting on its immediate surrounding only, it becomes necessary to add on top of this architecture, a distributed logic for decision-making based on the communication possibilities offered by the network and the computation power embedded in each node. This logic should allow the system of systems to apply uniform management rules and strategies to achieve its global objectives, to deal with potential conflicts between local decisions or their effects, and to pursue goals dedicated to some global optimization purposes.