基于演化算法的鲁棒性AUV分配方法研究

Research on Robust AUV Assignment Method Based on Evolutionary Algorithm

当前,海上监测以及水下可疑目标搜索方面的研究日益受到重视,各类水下搜索任务不断增多,复杂度日益提高,如何在限定的时间内科学合理高效地分配有限的搜索资源是其中的关键。针对此问题论文提出一种考虑不确定因素影响的改进模拟退火及遗传算法的AUV搜索分配方法。首先,分析AUV水下搜索任务情况并建立鲁棒模型,然后对模拟退火算法以及遗传算法的选择、交叉和变异等操作进行了改进,实现了对不确定情况下搜索资源分配方式的优化,最后通过Matlab仿真验证了该方法的可行性,有效地提高了搜索效率。

At present,research on marine monitoring and underwater suspicious target search is receiving increasing atten⁃tion.Various underwater search tasks are increasing and the complexity is increasing.The way to allocate limited search resources scientifically,rationally and efficiently within a limited time is the key.Aiming at this problem,this paper proposes an improved simulated annealing and genetic algorithm AUV search allocation method considering the influence of uncertain factors.First,the AUV underwater search task situation is analyzed and a robust model is established,and then the simulated annealing algorithm and the selection,crossover and mutation operations of the genetic algorithm are improved to achieve the optimization of the search re⁃source allocation method under uncertainty.Finally,the feasibility of this method is verified by Matlab simulation,and the search efficiency is effectively improved.