为了提高海流、海浪、水下噪声等扰动下的AUV三维轨迹跟踪精度，提出基于递归免疫网络在线辨识的PID自整定轨迹跟踪控制器（PID-RINN）。首先建立AUV的运动学模型，并将其深度距离、首向角和俯仰角作为控制变量，设计了基于神经网络在线辨识的PID控制器；然后借鉴生物免疫系统的信息处理机制构建了递归免疫网络；接着将水下机器人在水平面上距预瞄路径点的侧向距离定义为疫苗，并联合细胞隐层输出接种到递归免疫网络的突触隐层；最后基于梯度法实现了递归免疫网络辨识下的PID控制器在线自整定。测试结果表明，与PID、PID_ GA、PID _RBF相比，文中轨迹跟踪控制的平均和最大位置误差分别平均减少31.91%和25.81%，平均和最大首向角误差分别平均减少32.54%和25.27%，以及平均和最大俯仰角误差分别平均减少61.93%和61.26%，从而验证了文中递归免疫网络在线辨识的AUV三维轨迹跟踪具有高控制精度，以及强扰动抑制优点。
In order to improve the 3D trajectory tracking accuracy of autonomous underwater vehicle (AUV) under disturbances such as ocean current, ocean wave and underwater noise, a PID self-tuning trajectory tracking controller identified online by recurrent immune neural network (PID-RINN) is proposed. First, the kinematic model of the AUV is established, and the depth distance, heading angle and pitch angle of the AUV are taken as the control variables, and a PID controller identified online by the neural network is designed. Then a recursive immune network is constructed by referring to the information processing mechanism of the biological immune system. After that, the lateral distance between the underwater robot and the preview waypoint on the horizontal plane is defined as the vaccine, and is inoculated into the synaptic hidden layer of the recurrent immune network together with the output of the cellular hidden layer. Finally, based on the gradient method, the online self-tuning of the PID controller identified by the recursive immune network is realized. Test results show that compared with PID, GA_PID, RBF_PID, the average and maximum position errors of the proposed PID-RINN are respectively reduced by an average of 31.91% and 25.81%, the average and maximum heading angle errors are reduced by an average of 32.54% and 25.27% respectively, and the average and maximum pitch angle errors are reduced by an average of 61.93% and 61.26% respectively, which verifies that the 3D trajectory tracking identified online by the recursive immune network has the advantages of high control accuracy and strong disturbance suppression.
2024,46(13): 119-125 收稿日期：2023-09-04
DOI：10.3404/j.issn.1672-7649.2024.13.021
分类号：TP391.41;
基金项目：工信部高技术船舶项目（[2019]360）；张家港市产业链创新产品攻关计划资助项目（ZKC2206）；张家港市产学研预研资金资助项目（ZKYY2253）
作者简介：王舜(1999-),男,硕士研究生,研究方向为仿生智能、多机器人协作等
参考文献：
[1] 钟宏伟. 国外无人水下航行器装备与技术现状及展望[J]. 水下无人系统学报, 2017, 25(4): 215-225.
[2] 叶辉, 王丽琦, 智鹏飞, 等. 基于干扰观测器的AUV三维路径自适应跟踪控制[J]. 江苏科技大学学报(自然科学版), 2019, 33(5): 52-59.
[3] YUAN C, SHUAI C, MA J, et al. An efficient control allocation algorithm for over-actuated AUVs trajectory tracking with fault-tolerant control[J]. Ocean Engineering, 2023, 273(5): 113-976.
[4] 杜佳璐, 李健. 欠驱动水下机器人三维轨迹跟踪有限时间预设性能控制[J]. 控制理论与应用, 2022, 39(2): 383-392.
[5] SHOJAEI K, AREFI, MOHAMMAD M. On the neuro-adaptive feedback linearising control of underactuated autonomous underwater vehicles in three-dimensional space[J]. IET Control Theory and Applications, 2015, 9(8): 1264-1273.
[6] SHOJAEI K, DOLATSHAHI M. Line-of-sight target tracking control of underactuated autonomous underwater vehicles[J]. Ocean Engineering, 2017, 133(15): 244-252.
[7] LIU X, ZHANG M J, CHEN Z Y. Trajectory tracking control based on a virtual closed-loop system for autonomous underwater vehicles[J]. International Journal of Control, 2020, 93(12): 2789-2803.
[8] CAMPOS E, MONROY J, ABUNDIS H, et al. A nonlinear controller based on saturation functions with variable parameters to stabilize an AUV[J]. International Journal of Naval Architecture and Ocean Engineering, 2019, 11(1): 211-224.
[9] YAO F, YANG C, LIU X, et al. Experimental evaluation on depth control using improved model predictive control for autonomous underwater vehicle (AUVs)[J]. Sensors, 2018, 18(7): 20-22.
[10] 陈巍, 魏延辉, 曾建辉, 等. 基于干扰观测器的欠驱动AUV自适应反演控制[J]. 中南大学学报(自然科学版), 2017, 48(1): 69-76.
[11] 丁文俊, 张国宗, 刘海旻, 等. 面向海流扰动和通信时延的欠驱动AUV编队跟踪控制[J]. 兵工学报, 2023, 8(28): 1-13.
[12] 周则兴. 基于DRNN-S控制的AUV三维轨迹跟踪方法[J]. 舰船科学技术, 2021, 43(21): 96-99.
ZHOU Zexing. 3D trajectory tracking based on DRNN-S control for autonomous underwater vehicles[J]. Ship Science and Technology, 2021, 43(21): 96-99.
[13] 吴子明, 杨柯, 唐杨周, 等. 基于反步滑模控制的欠驱动AUV定深运动研究[J]. 舰船科学技术, 2023, 45(1): 114-119.
WU Ziming, YANG Ke, TANG Yangzhou, et al. Research on underactuated AUV depth motion based on backstepping sliding mode control[J]. Ship Science and Technology, 2023, 45(1): 114-119.
[14] YANG C, YAO F, ZHANG M J. Adaptive backstepping terminal sliding mode control method based on recurrent neural networks for autonomous underwater vehicle[J]. Chinese Journal of Mechanical Engineering, 2018, 31(1): 31-36.
[15] 王明阳. 基于模型预测的水下机器人养殖网箱巡检轨迹跟踪研究[D]. 大连: 大连海洋大学, 2022.
[16] 王通, 齐向东. 遗传算法PID控制在AUV姿态调节中的应用[J]. 自动化技术与应用, 2020, 39(4): 8-14.
[17] MUSTAFA W, NIZAR A. An adaptive neural network with nonlinear FOPID design of underwater robotic vehicle in the presence of disturbances, uncertainty and obstacles[J]. Ocean Engineering, 2023, 279(15): 29-36.
