A Practical Task Allocation Strategy Based on Parameter Adaptive Particle Swarm Optimization

Aiming at the task allocation problem of multi-UAV cooperative air combat system in battlefield environment, this paper proposes a task allocation strategy based on parameter adaptive particle swarm optimization algorithm (PAPSO). Firstly, the task allocation problem of UAV is described in the actual battlefield environment, and the mathematical model of task allocation is established based on the hit cost, damage cost and task revenue. Then, the correlation analysis method is used to assess the threat of the air combat situation of both enemy and us, and the obtained comprehensive threat matrix is used to calculate the task allocation model and serve as the basis of task allocation. Finally, a PAPSO algorithm based on evolutionary game theory is designed to solve the model. In this algorithm, a new state update formula is designed for each particle to prevent the particles from falling into iteration stagnation too early, and the evolutionary stability strategy in evolutionary game theory is introduced to adaptively update the control parameters of the algorithm, which balances the global and local search capabilities of the algorithm well, and a relaxation mechanism is also designed to reduce the loss of solution diversity in the early iteration. In this paper, a comparison experiment is designed to verify the proposed task allocation strategy. The simulation results illustrate that the proposed strategy in this paper has improved the quality of solutions and convergence speed compared to traditional strategies, which verifies its effectiveness and superiority.