针对存在不确定性的非线性船舶航向运动数学模型，将反步算法和command filter technique 相结合，提出一种新的非线性神经网络自适应控制器设计方法.基于Lyapunov稳定性分析理论，采用command filter technique消除传统反步法中存在的“计算爆炸”问题，同时将神经网络权值的范数作为估计参数，以减少控制器的计算负担，便于工程实现和应用.所设计的非线性自适应神经网络控制器使船舶航向控制系统中所有信号一致有界，航向误差收敛于零.仿真结果验证了该算法的有效性.
A new nonlinear adaptive control algorithm was proposed for course-keeping control of ship in the presence of uncertainties. This algorithm incorporated the command filter technique into Backstepping design, so the problem of “explosion of complexity” was avoided in the traditional Backstepping design procedure. In addition, the norms of neural network weights were used as the estimated parameters, and the computational burden of the algorithm can be reduced drastically and the algorithm is convenient to implement in applications. Result shows that the designed adaptive controller guarantees uniform boundedness of all signals in the coursekeeping control of ship and the heading error converges to zero. Simulation illustrates the effectiveness of the proposed algorithm.
［1］杜佳璐, 郭晨. 控制增益未知的船舶航向非线性自适应跟踪控制［J］. 控制理论与应用, 2005, 22(2):315-320.
DU Jia-lu, GUO Chen. Nonlinear adaptive design for coursetracking control of ship without a priori knowledge of control gain ［J］. Journal of Control Theory & Applications, 2005, 22(2):315-320.(in Chinese)
［2］罗伟林, 邹早建, 李铁山. 船舶航向非线性系统鲁棒跟踪控制［J］. 控制理论与应用, 2009, 26(8):893-895.
LUO Wei-lin, ZOU Zao-jian, LI Tie-shan. Robust tracking control of nonlinear ship steering ［J］. Journal of Control Theory and Applications, 2009, 26(8):893-895.（in Chinese）
［3］杨盐生. 船舶航向非线性系统的模型参考模糊自适应控制［J］. 中国造船, 2003, 44(3):85-93.
YANG Yan-sheng. Model reference fuzzy adaptive control for ship streering autopilot with uncertain nonlinear system ［J］. Shipbuilding of China, 2003, 44(3):85-93.(in Chinese)
［4］杨盐生. 不匹配不确定系统的变结构鲁棒控制及应用［J］. 交通运输工程学报, 2001, 1(2):102-107.
YANG Yan-sheng. Variable structure robust control algorithm for uncertain system with the absence of matching assumptions［J］. Journal of Traffic and Transportation Engineering, 2001, 1(2):102-107.(in Chinese)
［5］胡耀华, 贾欣乐. 广义预测控制应用于船舶航向和航迹保持［J］. 中国造船, 1998 (1):36-41.
HU Yao-hua, JIA Xin-le. The application of generalized predictive control to ship course and trackkeeping ［J］. Shipbuilding of China, 1998 (1):36-41.(in Chinese)
［6］贾欣乐, 张显库. H∞ 控制器应用于船舶自动舵［J］. 控制与决策, 1995, 10(3):250-254.
JIA Xin-le, ZHANG Xian-ku. H∞ controllers applied to autopilot for ships［J］. Control and Decision, 1995, 10(3):250-254.(in Chinese)
［7］LI Tie-shan, WANG Dan, FENG Gang, et al. A DSC approach to robust adaptive NN tracking control for strict-feedback nonlinear systems［J］ IEEE Transactions on Systems, Man, and Cybernetics-Part B: Cybernetics, 2010, 40(3):915-927.
［8］KWAN C, LEWIS F L. Robust Backstepping control of nonlinear systems using neural networks［J］ IEEE Transactions on Systems, Man, and CyberneticsPart A: Systems and Humans, 2000, 30(6):753-766.
［9］CHEN M, GE S S, HOW B V E. Robust adaptive neural network control for a class of uncertain MIMO nonlinear systems with input nonlinearities［J］. IEEE Transactions on Neural Networks, 2010, 21(5):796-812.
［10］CHEN B, LIU X P, LIU K F, et al. Direct adaptive fuzzy control of nonlinear strictfeedback systems［J］. Automatica, 2009, 45(6):1530-1535.
［11］FARRELL J A, POLYCARPOU M, SHARMA M, et al. Command filtered Backstepping ［J］. IEEE Transactions on Automatic Control, 2009, 54(6):1391-1395.
［12］DONG W J, FARRELL J A, POLYCARPOU M, et al. Command filtered adaptive Backstepping ［J］. IEEE Transactions on Control Systems Technology, 2012, 20(3):566-579.
