为了提高自主式潜航器（autonomous underwater vehicle, AUV）在作业期间的运动控制精度、鲁棒性能，提出一种融合HJI理论和递归神经网络的运动控制策略。考虑多种扰动因素建立动力学模型，引入对角递归神经网络实现系统中存在的多种不确定性和控制输入受限的有效补偿。以HJI不等式为基础设计鲁棒控制律，借助李雅普诺夫第二法证明设计的控制系统具有稳定性。仿真结果表明该方法的可行性和有效性，与对比方法相比，位置跟踪误差平均值减少50%以上，具有更高的控制精度、抗干扰能力，实现了对非线性轨迹的稳定跟踪控制。
To improve the motion control accuracy and robustness of an autonomous underwater vehicle (AUV) during operation, a motion control strategy combining HJI (Hamilton-Jacobi-Isaacs) theory and recursive neural network was proposed. The dynamic model was formulated by considering various disturbance factors, and the diagonal recursive neural network was introduced to realize the effective compensation of multiple uncertainties and control input constraints in the system. The robust control law is designed based on HJI inequality, and the stability of the designed controller was proved by the second method of Lyapunov. The simulation results show the feasibility and effectiveness of the method proposed in this paper. Compared with the comparison method, the position tracking error average is reduced by more than 50%. It has higher control precision and anti-interference ability, and realizes the stable tracking control of nonlinear trajectory.
2022,44(24): 100-106 收稿日期：2021-12-28
DOI：10.3404/j.issn.1672-7649.2022.24.021
分类号：U674.941
作者简介：陶健龙(1998-),男,硕士研究生,研究方向为自主式水下潜航器控制技术
参考文献：
[1] 曾晓光, 金伟晨, 赵羿羽, 等. 海洋开发装备技术发展现状与未来趋势研判[J]. 舰船科学技术, 2019, 41(9):1-7
[2] 李硕, 刘健, 徐会希, 等. 我国深海自主水下机器人的研究现状[J]. 中国科学:信息科学, 2018, 48(9):1152-1164
[3] 李福正. 导管螺旋桨匹配性及动态性能研究[D]. 哈尔滨:哈尔滨工业大学, 2017.
[4] DING H, LIU K P, YOU W L. Study on a new nonlinear H∞ guidance law for autonomous underwater vehicle[C]//Advanced Research on Information Science, Automation and Material System/Advanced Materials Research. 2011:219−220, 362−365.
[5] ISMAIL Z H, DUNNIGAN M W. Nonlinear H∞ optimal control scheme for an underwater vehicle with regional function formulation[J]. Journal of Applied Mathematics, 2013, 2013:732738
[6] MAHAPATRA S, SUBUDHI B. Nonlinear matrix inequality approach based heading control for an autonomous underwater vehicle with experimental realization[J]. IFAC Journal of Systems and Control, 2021, 16:100138
[7] MAHAPATRA S, SUBUDHI B, ROUT R, et al. Nonlinear H∞ control for an autonomous underwater vehicle in the vertical plane[J]. IFAC-PapersOnLine, 2016, 49(1):391-395
[8] DUAN K, FONG S, CHEN C L P. Reinforcement learning based model-free optimized trajectory tracking strategy design for an AUV[J]. Neurocomputing, 2022, 469:289-297
[9] DONG B, AN T, ZHU X, et al. Zero-sum game-based neuro-optimal control of modular robot manipulators with uncertain disturbance using critic only policy iteration[J]. Neurocomputing, 2021, 450:183-196
[10] WANG Y, SUN W, XIANG Y, et al. Neural network-based robust tracking control for robots[J]. Intelligent Automation and Soft Computing, 2009, 15(2):211-222
[11] KUMAR R, SRIVASTAVA S, GUPTA J R P. Diagonal recurrent neural network based adaptive control of nonlinear dynamical systems using lyapunov stability criterion[J]. ISA Transactions, 2017, 67:407-427
