Task Allocation in UAV Swarm Zone Breach: A Dynamic Game-Theoretical Approach

In this paper, the UAV swarm breach zone task allocation problem is modeled using potential games, incorporating the upper limit on UAVs per zone as a soft constraint in the utility function. Due to the dynamic nature of combat, the utility functions dynamically depend on the real-time positions of UAVs, making the model a dynamic game. Furthermore, an algorithm based on the generalized regret-matching mechanism (GRM) is proposed for dynamic games. Subsequently, we prove that the algorithm eventually converges to a Nash equilibrium closer to the terminal state. Finally, we demonstrate the advantages of the proposed algorithm compared to the traditional static game-solving approach in simulation.