基于DPSO-GT-SA算法的大规模UCAV协同任务分配
Cooperative Task Allocation of Large-Scale UCAV Based on DPSO-GT-SA Algorithm

无人作战飞机(UCAV)编队任务分配是研究UCAV编队飞行作战的关键．针对复杂约束环境下大规模UCAV协同任务分配问题, 提出改进离散粒子群算法．根据现有UCAV编队空对地饱和作战模式, 建立UCAV编队作战环境中任务分配模型, 通过采用离散粒子群优化-郭涛-模拟退火算法(DPSO-GT-SA)进行求解．根据粒子编码方式建立粒子与UCAV及目标之间的映射, 通过粒子交叉变异进行搜索与寻优, 并通过模拟退火Metropolis准则跳出局部最优．在复杂约束条件下, 为解决离散粒子群-郭涛算法(DPSO-GT)陷入局部极小问题, 引入改进模拟退火算法．为解决模拟退火后期收敛速度慢问题, 在DPSO-GT-SA算法中加入动态温度衰减因子．仿真结果表明, 改进离散粒子群算法可以更好地解决大规模UCAV协同任务分配问题．
Task allocation of unmanned combat aircraft vehicle(UCAV)is the key to UCAV formation flying campaign. An improved discrete particle swarm optimization(IDPSO)algorithm is proposed for large-scale UCAV cooperative task allocation in complex constraint environment. According to the existing multi-UCAV formation saturation combat mode of air to ground，the multiple UCAV formation task allocation model in combat environment is established. Discrete particle-GuoTao-simulated annealing(DPSO-GT-SA)algorithm is used to solve the problem. According to the particle coding method，the mapping between the UCAV，the target and the particle is established. The exploitation and exploration are based on crossover and mutation of particles，and the local optimum is jumped out by Metropolis criterion of simulated annealing. In order to solve the problem that the DPSO-GT algorithm falls into local minimum，the improved simulated annealing is introduced. As to the problem of slow convergence at the later stage of simulated annealing，a dynamic temperature attenuation factor in DPSO-GT-SA algorithm is proposed. The simulation results show that the improved discrete particle swarm optimization algorithm can better solve the large-scale UCAV cooperative task allocation problem