Game-theoretic multi-level methods for the control of large-scale systems

Project members:Dr. N.B. Groot (Noortje), B. De Schutter (Bart)
Keywords:Reverse Stackelberg games, Optimization-based control, Multi-level control, Intelligent control
Sponsored by:DCSC

One of the major challenges in the control of large-scale intelligent infrastructures such as road traffic networks is to find an efficient multi-level control scheme in which decisions by several controllers on different, mutually interacting control layers are taken into account.
This research project focuses on adopting game-theoretic elements in the modeling of communication and interaction between controllers on the various control levels. Here the trade-offs due to the multi-objective character of the problems have to be taken into account. Other extensions to consider are the different time scales on which controllers of different levels may operate and cases in which the leader does not have perfect information on e.g., the follower's actions.

A theory that seems to be particularly applicable to tackle these problems is the theory of reverse Stackelberg games. Reverse Stackelberg games are defined on the basis of a hierarchy between players; a player called the leader acts first by announcing her strategy as a mapping of the follower's decision space into her own decision space, while taking the response of the follower into account. The game is flexible in the sense that it may involve any number of leaders and followers, levels (i.e., some players have both a leader and a follower position towards lower and higher levels respectively), number of stages or time duration, etc.

Since the problem of finding an optimal strategy for the leader belongs to the realm of composed functions, these problems are known to be very difficult to solve in an analytic way. In this project we will therefore focus on the development of a solid framework for multi-level control based on the reverse Stackelberg game, which is moreover computationally efficient when applied to real-life problems.

© Copyright Delft Center for Systems and Control, Delft University of Technology, 2017.