基于改进的复合自适应遗传算法的UUV水下回收路径规划

Path Planning for UUV Underwater Recovery based on Improved Composite Adaptive Genetic Algorithm

传统遗传算法的变异操作会简单随机产生新的路径,对算法进化性能有不利影响,使算法易陷入局部最优的陷阱;遗传算法常配合栅格法进行路径规划,所得的最优路径并非无人水下航行器(UUV)回收路径规划可获得的最短路径,并存在UUV机动性能可能与最优路径冲突的问题。为此,设计一种具有UUV机动性约束条件的改进遗传算法,提出环境复杂度的概念用于分析机动性约束的具体取值,使路径规划适配于UUV的机动性,使算法结果更具实用性;提出复合自适应变异策略,控制变异的个体在迭代过程中发生自适应的进化;当一定迭代数内种群进化停滞时,引导最优个体进行双阶段自适应变异,从而使最优路径趋近全局近似最优解,有效提高算法的收敛速度。基于MATLAB软件的算法对比仿真结果表明一般复杂水域和复杂水域环境下,改进的复合自适应遗传算法生成的最优路径相比于遗传算法和自适应遗传算法的最优路径更加平滑,路径长度更低,可见改进的复合自适应遗传算法在路径规划上收敛性能和寻优能力更优,更具有可行性和优越性。

Mutations of traditional genetic algorithms generate new paths in a simple and random manner,which negatively influence the evolutionary performance of the algorithms and makes it easy for them to fall into the trap of local optimality.Moreover,genetic algorithms are usually used together with the grid method for path planning,and the optimal path obtained is not always the shortest path for UUV recovery path planning,and the UUV mobility performance might conflict with the optimal path.An improved genetic algorithm with UUV mobility constraints is thus proposed.The concept of environment complexity is proposed to analyze the specific value of mobility constraints,so that path planning can be adapted to UUV mobility,and the algorithm results can be more practical.The compound adaptive mutation strategy is proposed to control the adaptive evolution of the mutated individuals in the iterative process.When the population evolution stagnates after a certain number of iterations,the optimal individual is guided for a two-stage adaptive mutation so that the optimal path approaches the approximate global optimal solution,and the convergence rate of the algorithm is effectively improved.The algorithm comparison simulation results based on MATLAB software show that the optimal path generated by the improved compound adaptive genetic algorithm is smoother and shorter in length compared with the optimal path of genetic algorithm and adaptive genetic algorithm in generally complex water area and complex water area,which demonstrates that the improved compound adaptive genetic algorithm has better convergence performance and superiority seeking ability in path planning and is more feasible and superior.