大范围环境下自主式水下潜器两种全局路径规划方法的研究

应用遗传算法(GA)和A·算法对自主式水下潜器(简称AUV)在大范围海洋环境中的全局路径规划问题进行了研究.介绍了基于栅格的环境模型及其数据结构,讨论了GA的染色体编码方式、基于知识的初始种群生成方法与适应度函数,基于领域知识设计了五种遗传算子,给出了A·算法的具体实现方法.通过仿真结果可以看出:GA采用可变长编码方式使路径描述简单、清晰,具有收敛速度快、求解实际问题效率高的特点;A*算法可在较短时间内求得相对栅格优化的路径.两种算法均可满足系统实时性要求.

Research on Two Global Path Planning Methods for Autonomous Underwater Vehicle Based on Large-scale Chart Data

Global path planning problem for autonomous underwater vehicle (AUV) based on large-scale chart data is investigated by using two methods, which are Genetic Algorithm (GA) and A~* algorithm. This paper introduces environment model based on grid and data structure, in which the nodes of the grid store the digital elevation property. Some problems of GA, such as the coding of chromosome, generating initial population based on knowledge and evaluation function, and the design methods for five generic operators based on domain knowledge etc., are all discussed. And this paper also proposes the realization method for A~* algorithm. The simulation results show that GA makes the path described simply and clearly because of adopting a method of variable length codes, has the character of high speed global convergence, and can more efficiently solve the problem of path planning for AUV; the A~* algorithm can find a relative optimal path to grid in a little time; and both GA and A~* can satisfy the real time requirement for system.