基于增强学习控制器的仿生水下机器人姿态镇定研究  

Reinforcement Learning Controller Based Attitude Stabilization for Bionic Underwater Robots

作　　者：张卫武[1] 林龙信[2] 

机构地区：[1]西北工业大学航海学院,西安710072 [2]海军装备研究院,北京100161

出　　处：《计算机测量与控制》2012年第11期3063-3065,共3页Computer Measurement &Control

摘　　要：仿生水下机器人是水下机器人领域的一个重要研究方向;利用增强学习控制器对仿生水下机器人的姿态镇定问题进行了研究;增强学习控制器主要由回报函数、学习样本数据库、神经网络、动作选择以及Q学习算法等模块构成,可通过直接与环境交互生成最优动作选择策略;镇定仿生水下机器人的偏航角姿态镇定的仿真试验表明,增强学习控制器在偏航角姿态镇定方面的性能较为理想;学习样本数据库的引入显著提升了增强学习控制器的姿态镇定性能;学习样本数据库的容量对学习性能存在较大影响。The bionic underwater robot is one of the most important fields in underwater robots. This paper discussed the attitude stabili- zation problem of bionic underwater robots based on the reinforcement learning controller which is composed of reward function, database of learning samples, neural network, action selection and the Q--learning algorithm. Simulation experiments about the yawing angle stabiliza- tion are carried out, and results indicate that the reinforcement learning controller has a good performance in yawing angle stabilization; the database of learning samples improves the attitude stabilization performance evidently; the capacity of the database of learning samples has a big influence on the learning performance.

关 键 词：增强学习控制器 仿生水下机器人 姿态镇定 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]