基于差分-灰狼算法的水下航行器路径规划方法

A differential grey wolf optimizer for underwater vehicles path planning

水下航行器因其自主灵活、安全可靠的特性已被日趋广泛地应用于各类海洋探测任务中。然而,其路径规划仍面临着算法求解效果不佳、路径安全性不足等问题。本文以路径安全性为约束条件,路径总长度最小化为目标函数建立了水下航行器路径规划模型。针对该模型的求解需求,本文将差分进化算子引入灰狼算法中以改善其全局探索能力,并称改进算法为差分-灰狼算法。仿真实验结果表明,本文所提出的差分-灰狼算法可以为水下航行器规划安全经济的路径,并在规划效果与收敛速度方面较差分进化算法和灰狼算法有明显优势,具有广阔工程应用价值。

Underwater vehicles have been widely used in various ocean exploration missions due to their autonomous,flexible,safe,and reliable characteristics.However,their path planning still face problems such as poor performance and insufficient security.This article establishes an underwater vehicle path planning model with path safety as the constraint and minimizing the total length of the path as the objective function.In response to the solving requirements of the model,this article introduces the differential evolution operator into the grey wolf optimizer to improve its global exploration ability and calls the improved algorithm the differential grey wolf optimizer(DGWO).The simulation results show that the proposed DGWO can plan safe and economic paths for the underwater vehicle.Additionally,the DGWO not only has obvious advantages in planning solutions and convergence rates but also has broad engineering application value.