摘要
以能量消耗量作为评价函数进行水下机器人运动规划的结果得到随时间变化的目标航向角和理论控制输出 .本文探讨基于神经网络的追踪控制器以实现在运动控制过程中对水下机器人被规划航向角进行追踪控制 ,并探讨实际控制输出与理论控制输出的一致性问题以实现运动过程消耗的操作能量较小 .水下机器人“Twin -Burger”
On the basis of the results of the AUV(Autonomous Underwater Vehicle) motion planning by supposing the operating energy as evaluation function, the objective heading angle and theoretical control output varied with time are obtained. In this paper, a trail controller based on neural network is given to objective path trail in the motion control, and the conformance problem of the random values of the practical and theoretical control output is discussed to realize the motion process with less energy consumption. The result of the experiment in pool by AUV “Twin-Burger” proves that the performance methods presented in this paper are reasonable.
出处
《哈尔滨工程大学学报》
EI
CAS
CSCD
1999年第6期1-8, ,共8页
Journal of Harbin Engineering University
