CASIA OpenIR  > 毕业生  > 博士学位论文
仿生机器鱼信息处理和控制研究
Alternative TitleResearch on Information Processing and Control of Biomimetic Robot Fish
沈志忠
Subtype工学博士
Thesis Advisor谭民 ; 王硕
2006-06-05
Degree Grantor中国科学院研究生院
Place of Conferral中国科学院自动化研究所
Degree Discipline控制理论与控制工程
Keyword仿生机器鱼 运动规划 视觉跟踪 水下监控 Biomimetic Robot Fish Motion Planning Visual Tracking Underwater Monitoring
Abstract仿生机器鱼作为一种新型水下潜器,具有广阔的应用前景,本文主要针对仿生机器鱼的信息处理和控制展开研究,从底层控制、高层规划、协调作业到水下传感网络系统节点设计等不同层次研究仿生机器鱼控制、信息处理与未来应用等问题,文中主要给出了以下一些研究结果: 本文首先对仿生机器鱼的应用背景和研究意义进行了介绍,综述了国内外在鱼类推进机理、仿生机器鱼研制以及控制方面的研究现状,介绍了本文的选题背景和主要研究内容。 第二,研究了机器鱼的底层反应式自主避障功能,给出了集成超声和红外传感器的仿生机器鱼系统设计方案,提出了一种基于增强式学习的仿生机器鱼自主避障策略,给出了状态集和行为集,采用当场奖励和延时奖励相结合的方法,通过学习获得了有效的状态-行为组合。 第三,在恒定阈值分割方法的基础上,提出了一种自适应阈值图像分割方法,改善了基于固定阈值的图像分割算法不能适应环境光照变化的问题,并应用此方法来进行仿生机器鱼的视觉信息处理。 第四,基于视觉获取的姿态反馈信息,提出了一种基于Bezier曲线的路径生成和运动控制算法,从机器鱼的当前状态到期望目标状态规划出一条平滑的Bezier曲线,然后将该曲线路径离散化为多个点,通过方向控制、速度控制实现机器鱼沿规划路径运动,并在机器鱼完成射门任务的实验中验证了该算法的有效性。 第五,提出了一种多机器鱼协作进行目标搜索的算法。结合在一定范围内搜索热源的任务,基于多机器鱼编队技术,给出了机器鱼队形形成算法和目标搜索算法。 第六,提出了一种以机器鱼为节点的移动传感器网络体系结构,并初步实现了由机器鱼子系统、通讯子系统和地面控制台三部分组成的水下监控系统。 最后,对所开展的工作进行了总结,并指出了需要进一步研究的工作。
Other AbstractBiomimetic robot fish, as a new style of submarine thruster, can play an important role in many applications. Research on information processing and control of the robot fish is mainly carried out in this paper. Biomimetic robot fish’s control, information processing and future applications are studied in different layers, such as bottom control, top planning, coorperation and node design of underwater sensor networks, etc. Researc productions are as follows. Firstly, the application background and research purposes of biomimetic robot fish are introduced. The research trend and progress in fish’s propulsive mechanism, robot fish’s research, development and control in the world are reviewed. The background and structure of the thesis are also given. Secondly, autonomous reactive control for obstacle avoidance of the robot fish is studied. The design of a biomimetic robot fish system with ultrasonic and infrared sensors is given. An obstacle avoidance strategy based on reinforcement learning is proposed in this paper. The state and behavior sets are designed. Through adopting immediate and delayed rewards, valid state-behavior pairs are gotten through learning. Thirdly, an adaptive image segmentation algorithm is put forward in this paper. The proposed algorithm can improve the segmentation performance, and solve the problem that image segmentation algorithm based on fixed thresholding cannot suit the light variation. The algorithm is applied to the auxiliary visual system for the biomimetic robot fish control. Fourthly, based on visual information, a kind of Bezier curve-based path generation and motion control algorithm for the biomimetic robot fish is proposed in the paper. A smooth Bezier curve can be planned from robot fish current state to the expected state. Then multiple discrete points are gained from the curve. Robot fish’s locomotion along the planned path is realized through direction and velocity control. The validity of the algorithm is tested through a shooting gate experiment. Fifthly, coorperation algorithm of multiple robot fishes for object searching is proposed. Focusing on a practical problem, finding the heat source position in finite water area, the algorithms of forming equilateral triangle and searching based on formation technique are presented to locate the heat source position. Sixthly, the architecture for underwater mobile sensor networks is presented in this paper. The robot fish acts as the node in this architecture. An underwater monitoring system based on a robot fish is built up. This system is composed of robot fish subsystem, communication subsystem and controller. Finally, the obtained results are summarized and future work is addressed.
shelfnumXWLW1018
Other Identifier200318014602982
Language中文
Document Type学位论文
Identifierhttp://ir.ia.ac.cn/handle/173211/5938
Collection毕业生_博士学位论文
Recommended Citation
GB/T 7714
沈志忠. 仿生机器鱼信息处理和控制研究[D]. 中国科学院自动化研究所. 中国科学院研究生院,2006.
Files in This Item:
File Name/Size DocType Version Access License
CASIA_20031801460298(2829KB) 暂不开放CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[沈志忠]'s Articles
Baidu academic
Similar articles in Baidu academic
[沈志忠]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[沈志忠]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.