基于动态神经网络的AUV航向自适应控制

海洋环境的复杂性以及自身模型的不确定性,给自主潜航器(AUV)航向控制带来很大困难.针对AUV的特点和控制方面所存在的问题,采用了带衡量因子的动态BP神经网络控制器控制AUV的航向.理论分析和仿真结果表明,与传统的PID控制器相比,在扰动存在的情况下,神经网络控制器具有更好的自适应性和鲁棒性.

AUV heading adaptive control based on the dynamic neural network

The complexity of sea environment and the uncertainty of AUV model lead to difficulties of AUV heading control. For the existing issues in the features and control of AUV,it adopts dynamic BP neural network with evaluating factors to control the AUV heading control. The theoretical analysis and simulating results show that,comparing with the traditional PID controller,the controller of the neural network has better self-adaptability and robustness,under the condition of existing turbulence.