基于RBF神经网络的AUV自适应动态积分滑模轨迹跟踪  

Adaptive Dynamic Integral Sliding Mode Trajectory Tracking of AUV Based on RBF Neural Network

作　　者：马洪潮 戴晓强[1] 陆震 曾庆军[1] 许赫威 MA Hongchao;DAI Xiaoqiang;LU Zhen;ZENG Qingjun;XU Hewei(School of Telecommunications,Jiangsu University of Science and Technology,Zhenjiang 212028,China;No.704 Research Institute of China Shipbuilding Industry Group,Shanghai 200031,China)

机构地区：[1]江苏科技大学电子信息学院,江苏镇江212028 [2]中国船舶重工集团第七〇四研究所,上海200031

出　　处：《火力与指挥控制》2022年第12期65-71,共7页Fire Control & Command Control

基　　金：江苏省产业前瞻与共性关键技术基金资助项目(BE2018103)。

摘　　要：为解决欠驱动AUV在海流干扰、控制器超调及模型不确定下的三维轨迹跟踪控制问题,在运动学控制器中,设计具有自适应律的虚拟向导,减弱了洋流干扰提高了系统稳定性。在动力学控制器中,引入二阶微分跟踪器,确保纵向速度控制过渡时间短、不出现超调;并应用RBF神经网络对AUV模型的不确定项进行逼近并补偿。利用李亚普诺夫稳定性理论证明整个控制器的稳定性。仿真结果表明:该控制器能够有效克服洋流和模型不确定的影响,抑制抖振和克服超调,实现对三维轨迹的跟踪。In order to solve the three-dimensional trajectory tracking control problem of underactuated AUV under current disturbance,controller overshoot and model uncertainty,a virtual guide with adaptive law is designed in the kinematics controller to reduce the current disturbance and to improve the system stability.In the dynamic controller,a second-order differential tracker is introduced to ensure that the transition time of longitudinal speed control is short and there is no overshoot;RBF neural network is used to approximate and compensate the uncertainty of AUV model.Finally,Lyapunov stability theory is used to prove the stability of the whole controller.The simulation results show that the controller can effectively overcome the influence of ocean current and model uncertainty,and can suppress chattering and overcome to overshoot,to realize the tracking of three-dimensional trajectory.

关 键 词：AUV 轨迹跟踪 RBF神经网络 二阶微分跟踪器 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]