欠驱动AUV自适应神经网络反步滑模跟踪控制

Path following of an underactuated AUV based on adaptive neural network backstepping sliding mode control

针对存在参数不确定性和外界未知干扰的欠驱动自主水下航行器(AUV)三维路径跟踪问题,提出一种基于神经网络的反步滑模控制策略.首先,利用虚拟向导的方法建立了欠驱动AUV三维路径跟踪误差模型;其次,基于李雅普诺夫稳定性理论,利用反步法和滑模控制方法设计一种自适应鲁棒控制器,并设计一种在线调节增益切换函数以降低系统抖振,同时采用径向基函数(RBF)神经网络控制技术对AUV系统中不确定参数以及外界非线性千扰进行自适应补偿估计,而后利用李雅普诺夫稳定性理论证明了整个闭环系统的稳定性;最后,针对一种新型飞翼式欠驱动AUV进行数值仿真实验,结果表明所设计控制器可以实现对三维路径的精确跟踪,并对外界非线性干扰具有良好的鲁棒性.

An adaptive neural network backstepping sliding mode control scheme was presented in order to address the problem of the three-dimensional path following for the underactuated autonomous underwater vehicle(AUV)with the parameter uncertainties and external disturbances.Firstly,the three-dimensional path following error model was established by the virtual guidance method.Then,based on the Lyapunov stability theory,an adaptive robust control scheme was presented by backstepping and sliding mode control method,and an online adjustment gain switching function was designed to reduce the control system chattering.The radial basis function(RBF)neural network control method was introduced to carry out the unknown nonlinear uncertainties and external disturbances of the underactuated AUV,and the stability of the whole closed loop system was demonstrated by the Lyapunov stability theory.Finally,the simulation results verify the high accuracy of the three-dimensional tracking and good robustness of the proposed control scheme through a new class of flying wing autonomous underwater vehicle.