基于FNN优化的AUV姿态控制研究  

Research on AUV recovery attitude control based on improved active disturbance rejection control

作　　者：张海龙[1] 齐向东 普勇博 张涛[1] ZHANG Hailong;QI Xiangdong;PU Yongbo;ZHANG Tao(Key Laboratory of Instrumentation Science and Dynamic Measurement of Ministry of Education North University of China,Taiyuan 030051,China)

机构地区：[1]中北大学仪器科学与动态测试教育部重点实验室,山西太原030051

出　　处：《舰船科学技术》2025年第5期132-137,共6页Ship Science and Technology

基　　金：山西省科技创新团队专项资助项目(202304051001030);水声对抗技术国防科技重点实验室基金项目(2023JCJQLB3302)。

摘　　要：为了满足自主水下潜航器(AUV)快速达到稳定姿态的需求,在传统增量式PID的基础上引入神经网络理论和模糊控制逻辑,提出一种模糊神经网络(FNN)PID姿态控制器。首先建立双坐标系系统,并通过受力分析得到AUV动力学模型,其次融合模糊逻辑和人工神经网络的计算模型,设计AUV姿态控制器并搭建Matlab仿真模型,有效解决模糊PID控制过度依赖经验,难以应对水下复杂工况的问题。仿真结果表明,相比于传统的模糊PID控制和BP神经网络,模糊神经网络PID姿态控制器具有更快的响应速度,达到稳定姿态所需时间减少一倍以上,有效改善了AUV姿态控制性能。In order to meet the needs of autonomous underwater vehicles(AUV)to quickly achieve stable attitudes,a fuzzy neural network(FNN)PID attitude controller was proposed by introducing neural network theory and fuzzy control theory on the basis of traditional incremental PID.Firstly,the dual coordinate system is established,and the dynamics model of the underwater vehicle is obtained by force analysis,and then the fuzzy logic and artificial neural network are combined,the AUV attitude controller is designed and the Matlab simulation model is built,which effectively solves the problem that fuzzy PID control is overly dependent on experience and difficult to cope with complex underwater conditions.The simula-tion results show that compared with the traditional fuzzy PID control and BP neural network,The simulation results show that compared with the traditional fuzzy PID control and BP neural network,the fuzzy neural network PID attitude control-ler has better response speed,and the time required to achieve stable attitude is more than doubled,which effectively im-proves the AUV attitude control performance.

关 键 词：自主水下潜航器 模糊PID BP神经网络 控制优化 

分 类 号：U674.91[交通运输工程—船舶及航道工程]