Evolution modeling and control of networked dynamic games with event-triggering mechanism.

Area of Science:

• Artificial Intelligence
• Game Theory
• Network Science

Background:

• Decision-making in dynamic environments with multiple intelligent agents presents complex challenges.
• Existing models may not fully capture the temporal and strategic evolution of interactions.

Purpose of the Study:

• To introduce and analyze event dynamic games (EDGs) for discrete event decision-making.
• To develop novel state space and network evolution models for EDGs.
• To investigate the equilibrium solutions and temporal evolution behaviors within EDGs.

Main Methods:

• Development of a novel state space model for EDGs.
• Application of network evolution principles to model EDG networks.
• Establishment of a game tree evolution model using a normal-form strategic approach.
• Analysis of multi-layered temporal evolution and cooperative node behaviors.

Main Results:

• Introduction of a novel state space model for EDGs with analysis of equilibrium existence.
• Development of a game tree evolution model that extends strategic expression.
• Creation of a network evolution model to analyze temporal dynamics and cooperative behaviors.
• Validation of the proposed models through application examples.

Conclusions:

• The proposed models effectively capture the discrete event dynamic decision-making process in noncooperative settings.
• The framework provides insights into the evolution of strategies and cooperative behaviors in dynamic game networks.
• The study offers a robust method for analyzing complex interactions in multi-agent systems.