设为首页 联系我们 加入收藏

当前位置: 网站首页 期刊分类目录 2025第6期 正文

基于RBF神经网络的波浪滑翔器航向控制

作者:时间:2025-12-28点击数:



全文下载: 202506016.pdf


文章编号: 1672-6987(2025)06-0126-09 DOI: 10.16351/j.1672-6987.2025.06.016

崔萌, 杨燕*(青岛科技大学 自动化与电子工程学院, 山东 青岛 266061)

摘要: 波浪滑翔器作为一种新型海洋无人观测平台,面对海洋环境中强非线性、多扰动等复杂因素,传统PID控制方法在航向控制中的适应性和控制效果受到限制。为解决该问题,提出一种融合RBF径向基神经网络的增量式PID控制策略。首先构建波浪滑翔器动力学模型,其次设计RBF-PID控制器并给出推导公式,该策略利用RBF神经网络对增量式PID的3个参数进行自适应调整,确保实际航向快速逼近期望值。为进一步降低计算量,引入遗传算法优化网络初始权值。最终通过仿真平台与海试对所提算法进行了验证,结果表明该算法在复杂海况下依然具备良好的航向跟踪能力,显著提升了航向稳定性与控制精度。


关键词: 波浪滑翔器; RBF神经网络; 航向控制; 遗传算法


中图分类号: TM 46        文献标志码: A


引用格式: 崔萌, 杨燕. 基于RBF神经网络的波浪滑翔器航向控制[J]. 青岛科技大学学报(自然科学版), 2025, 46(6): 126-134.



CUI Meng, YANG Yan. Wave glider heading control based on rbf neural network[J]. Journal of Qingdao University of Science and Technology(Natural Science Edition), 2025, 46(6): -.

Wave Glider Heading Control Based on RBF Neural Network


CUI Meng, YANG YanCollege of Automation and Electronic Engineering, Qingdao University of Science and Technology, Qingdao 266061, China)

Abstract: As a new type of unmanned Marine observation platform, the wave glider is confronted with complex factors such as strong nonlinearity and multiple disturbances in the Marine environment. The adaptability and control effect of the traditional PID control method in heading control are limited. To solve this problem, an incremental PID control strategy integrating the RBF radial basis neural network is proposed. Firstly, the dynamic model of the wave glider is constructed. Secondly, the RBF-PID controller is designed and the derivation formula is given. This strategy utilizes the RBF neural network to adaptively adjust the three parameters of the incremental PID to ensure that the actual heading rapidly approaches the expected value. To further reduce the computational load, the genetic algorithm is introduced to optimize the initial weights of the network. Finally, the proposed algorithm was verified through the simulation platform and sea trials. The results show that the algorithm still has a good heading tracking ability under complex sea conditions, significantly improving the heading stability and control accuracy.


Key words: wave glider; radial basis neural network; heading control; genetic algorithm

收稿日期: 2024-12-12

基金项目: 山东省重大科技创新工程项目(2019JZZY020701); 国家重点研发计划项目 (SQ2020YFF0426588).

作者简介: 崔萌(1996—), 男, 硕士研究生.     * 通信联系人.


Copyright © 2011-2017 青岛科技大学学报 (自然科学版)