基于模糊RBF神经网络PID的AUV姿态控制研究  

Research on AUV attitude control based on fuzzy RBF neural network PID

作　　者：牛亮 党晓圆 冯元 崔卫星 NIU Liang;DANG Xiaoyuan;FENG Yuan;CUI Weixing(Chongqing College of Mobile Communication,Chongqing 401520,China)

机构地区：[1]重庆移通学院,重庆401520

出　　处：《传感器与微系统》2024年第10期11-14,共4页Transducer and Microsystem Technologies

基　　金：重庆市教育委员会科学技术研究项目(KJQN202302405)。

摘　　要：针对自主水下航行器(AUV)高精度、强鲁棒性的运动姿态控制需求,提出了一种径向基函数(RBF)神经网络结合模糊PID控制的水下机器人运动控制器;采用RBF神经网络对模糊PID控制器参数进行优化,有效解决了模糊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:模糊RBF神经网络PID控制器在AUV姿态调节中表现出较传统模糊PID控制器更好的响应速度和抗干扰能力,有效改善了AUV姿态控制性能;经实际应用验证,控制器在复杂工况下可以快速收敛至期望姿态并维持稳定。Aiming at the motion attitute control requirement of high precision and strong robustness of autonomous underwater vehicle(AUV),a motion controller of AUV based on radial basis function(RBF)neural network combined with fuzzy PID control is proposed.The RBF neural network is used to optimize the parameters of fuzzy PID controller,which effectively solves the problem that fuzzy PID control relies too much on experience and is difficult to deal with complex underwater working conditions.The simulation results show that the fuzzy RBF neural network PID controller has better response speed and anti-interference ability than the traditional fuzzy PID controller in AUV attitude adjustment,and effectively improves the AUV attitude control performance.The practical application proves that the controller can quickly converge to the desired attitude and maintain stability under complex working conditions.

关 键 词：自主水下航行器 运动控制 径向基函数神经网络 模糊PID 运动控制器 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]