摘要
本文介绍一种基于神经网络的水下智能潜器的运动控制方法，该方法通过在线学习，融控制与滤波为一体。计算机仿真与水池实验验证表明，该方法的控制与滤波性能良好，对环境的学习与适应能力强。该方法事实上可用于一般动力系统的控制。
This paper introduces an underwater vehicle motion control method which involves a learning control strategy based on neural networks. Through on-line learning,such method can blend control and filtering as a whole. The results of computer simulations and tank exper iments show that the performance of the control and filtering is good and that the learning and self adaptive ability is strong. Actually this method can be used for the control of a general dynamic system.
出处
《海洋工程》
CSCD
北大核心
1995年第2期38-46,共9页
The Ocean Engineering
关键词
智能潜器
学习控制
神经网络
自适应滤波
intelligent underwater vehicle
learning control
neural networks
selfadaptive filtering
