基于双重遗传算法的多机器鱼路径规划

为更快地计算出机器鱼之间在协调合作下的最优路径,提出一种基于改进的遗传算法的多目标路径规划方法。在传统遗传算法的基础上加入了插入、删除和修复算子,提高了算法的搜索效率,在选择算子中加入了避免外部存储器中出现相同个体的机制,防止＂早熟＂收敛,并将该算法引入到多机器鱼路径规划中,通过变异、选择等操作得到最优路径。水中机器鱼比赛2D仿真平台上的实验结果表明：该算法具有较快的搜索效率和较强的适应性,并大大提高了系统的协调性。

Multi-Robot Fish Path Planning Based on Double-Layer Genetic Algorithm

For faster to calculate the optimal path among the robotic fishes which work at the condition of cooperation mechanism, a multi-objective path planning method based on the improved genetic algorithm is proposed. For improving the search efficiency, the inserting, deleting and repairing operator are introduced in the method. To avoid the premature convergence, the mechanism which prevents appearing the same unit in external memorizer is added into the selection operator. Applying the method to the path planning of multi-robot fishes, the optimal path is obtained through mutation,selection and other operations. The experimental results on the 2D simulation platform show that the algorithm has faster searching efficiency and better adaptability, and improves the coordination of the system greatly.