复杂网络能控性鲁棒性研究进展

The study of controllability robustness is valuable to the control and application of various complex systems, including social, biological, and technological networks. Many real-world systems can be modeled as complex networks. The concept of controllability of complex networks refers to the ability of a network being steered by external inputs from any of its initial state to any desired target state under an admissible control input within a finite duration of time. The controllability robustness reflects how well the system can maintain the controllability against malicious attacks by means of node removals or edge removals. This survey gives a systematic investigation in the recent progress of the controllability robustness of complex networks. Firstly, the definitions and measures of controllability robustness are introduced. Then, the controllability robustness is considered from the perspective of attacks. Three types of attack strategies are discussed, including random attacks, feature-based targeted attacks, and heuristic-based attacks. Optimization methods toward stronger controllability robustness are investigated, including network modeling, edge rewiring, etc. Recent progresses have been achieved in both effective attack strategies and efficient topological optimizations, which provide a basis for further theoretical analysis. Finally, some potential future works are suggested.