基于遗传蚂蚁混合算法的AUV全局路径规划

Global path planning based on genetic-ant hybrid algorithm for AUV

针对三维空间环境下的全局路径规划问题存在搜索最优解效率不高、时间较长等问题,提出了一种基于遗传蚂蚁混合的算法.该算法对模型进行了必要的简化,合理设计了子区域的划分方法,减少了种群的搜索范围;采用了新的遗传蚂蚁混合策略,提高了两种基本算法的融合效率;依据安全性和时间原则设计了算法的评价函数.三维环境下的仿真结果表明:本算法与已有的遗传蚂蚁混合算法相比,在保持搜索精度的同时,搜索到最优解的时间和迭代次数均有明显减小,具有可行性与合理性.

Aiming at the problem of path planning in three-dimensional space environment, such as the low efficiency of searching optimal solution and the long searching time, a genetic-ant hybrid algo- rithm was proposed. With the necessary simplification of the model, rational sub region division method was designed to reduce the population searching range. Then, an optimal integration strategy was put forward to improve the integration efficiency, and the evaluation function was set based on the principle of the safety and the time. Simulation experiments in three-dimensional environment show that the proposed algorithm has the advantage that the search time and the number of iterations are obviously reduced while the search precision kept compared with the existing hybrid algorithms, which is feasible and reasonable.