一种具备协作机制的市场化无人战斗机任务分配方法

A Market-based Task Allocation Method with Cooperative Mechanism for Multi-UCAV Cooperative Mission Planning

当多无人战斗机编队需要对多个目标的敌方阵地进行攻击时,如何在有限的无人战斗机中选择最优的任务分配方案是UCAV编队能够有效完成任务的关键。首先提出了的方案优劣判定定理,然后将目标看作优化对象,根据UCAV的数目和目标的数目之间的差别制定了相应的任务分配策略。其次,基于相应的任务分配策略提出了具备合作机制的市场化的任务分配方法。该方法通过市场上买卖的方法和优化目标之间的相互协作,可以求出满足帕累托优化条件的最优任务分配方案。最后将本方法和基于粒子群的任务分配方法进行了对比分析并且进行了仿真验证。仿真结果表明使用本方法可以消除使用粒子群算法给帕累托最优方案的求解带来的不确定性。在局部代价和收益发生变化时,通过局部协调就可以获得最佳任务分配方案。

When multi-uninhabited combat air vehicle(UCAV) formation attacks the enemy's targets in the battle field,how to acquire the optimal task allocation result among finite UCAVs is the key problem that UCAVs are able to complete their missions effectively.Firstly,this paper presented a task allocation criterion.It assumed the selections of the targets as the optimized objectives and introduced the task allocation strategies according to the relationships of the number of UCAVs and that of targets.Secondly,it presented a novel market-based task allocation method with cooperative mechanism according to the allocation strategies.This method utilizes the 'buy' and 'sell' operations in a trade and adjusts the allocation results with cooperative mechanism among a number of targets and can acquire the Pareto optimized task allocation result.Finally,it compared our method with the Particle Swarm optimization(PSO)-based task allocation method and simulated them. Simulation shows that using our method eliminates the uncertainty of using the PSO algorithm,and when the local cost and value were changed,the optimal task allocation result can be obtained according to the local cooperation.