CASIA OpenIR  > 毕业生  > 博士学位论文
仿生游动机器的建模与智能控制方法研究
Alternative TitleModeling and Intelligent Control for Underwater Biomimetic Vehicle
尚留记
Subtype工学博士
Thesis Advisor谭民 ; 王硕
2011-05-24
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仿生游动机器 波动推进 建模 模糊-pid控制 路径规划 Underwater Biomimetic Vehicle Undulatory Propulsion Modeling Fuzzy Logic Pid Control Path Planning
Abstract水下仿生游动机器具有机动性好、推进效率高等优点,应用前景广阔。而两侧波动鳍推进型仿生游动机器是一类典型的水下仿生游动机器系统。本文针对两侧波动鳍推进型仿生游动机器,重点开展了其系统设计、建模、控制和路径规划等方面的研究工作,论文的主要内容如下: 首先,基于仿生机理和系统控制要求,给出了波动鳍推进游动机器机构设计、基于FPGA的游动机器驱动系统设计、基于ARM-Linux的实时控制系统设计,以及基于惯导系统的传感器系统设计。在此基础上,通过系统集成开发了游动机器的实验平台,以及辅助游动机器实验研究的远程控制平台。 其次,通过分析波动鳍的运动特征,提出并验证了波动鳍的运动控制参数调节方法。分析并提出了两侧波动鳍的协调控制策略,实现了游动机器前向、后向、侧向、旋转等多种游动模式。在此基础上,提出了基于有限状态机和游动模式切换相结合的避障控制策略。 第三,开展多种游动模式的开环控制实验研究,通过实验数据分析提出了一种合理的游动机器模型简化方法,进而结合加权递归最小二乘、神经网络等方法建立了几类典型游动模式下游动机器控制量和输出量之间的模型。 第四,提出了基于模糊控制器和PID控制器相结合的切换控制方法,实现了游动机器偏航角和速度的控制。进一步改进游动机器系统控制结构,在模糊-PID切换控制基础上增加了基于参考模型的前馈补偿,提升了系统响应速度。 第五,深入分析了游动机器的路径规划要求和机构约束,提出了基于贝塞尔曲线的游动机器三维空间可行路径规划算法。在此基础上,针对在线规划、重规划和避障等要求进一步完善了路径规划算法。 最后,对本文工作进行了总结,并指出了需要进一步开展的研究工作。
Other AbstractUnderwater biomimetic vehicle plays an important role in many practical applications due to its higher efficiency and better maneuverability. Furthermore, the vehicle propelled by two undulatory fins is a typical underwater biomimetic vehicle. This thesis focuses on the system design, modeling, control and path planning of the vehicle propelled by two undulatory fins. The main content of this thesis is described as follows. Firstly, based on the biomimetic mechanism and system control requirements, we designed a prototype of swimming vehicle with two undulatory fins, an FPGA-based driving system, an ARM-Linux-based real-time control system, and a sensor system based on inertial units, respectively. Then, by integrating the above elements, an experimental system and a remote control platform were designed. Secondly, by studying the motion characteristics of undulatory fins, an adjusting algorithm for the control parameters of the undulatory fin is proposed. By considering the coordination control methods of two undulatory fins, several swimming motion modes are developed, such as marching mode, receding mode, rotating mode, side-swaying motion mode etc. On the basis of these motion modes, an obstacle avoidance strategy is proposed by integrating the swimming motion mode switch and the finite-automata approach. Thirdly, by considerable open-loop swimming control experiments, a reasonable model simplification method is proposed based on experimental data analysis. Then by recursive weighted least square method and neural network approach, several input-output models were developed for the typical swimming motion modes. Fourthly, a yaw angle and speed control method for the vehicle was proposed by combining fuzzy control and PID control. In addition, by improving the controller structure of vehicle, a model-based feed-forward compensation control algorithm is proposed to increase the system responding speed. Fifthly, a Bezier curve based 3D path planning algorithm is proposed by considering the path planning requirements and mechanical constraints of the vehicle. Moreover, according to the requirements of online planning, secondary planning and obstacle avoidance, the proposed path planning algorithm is improved. Finally, the whole thesis is summarized and future work is addressed.
shelfnumXWLW1574
Other Identifier200818014628016
Language中文
Document Type学位论文
Identifierhttp://ir.ia.ac.cn/handle/173211/6333
Collection毕业生_博士学位论文
Recommended Citation
GB/T 7714
尚留记. 仿生游动机器的建模与智能控制方法研究[D]. 中国科学院自动化研究所. 中国科学院研究生院,2011.
Files in This Item:
File Name/Size DocType Version Access License
CASIA_20081801462801(3173KB) 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
[尚留记]'s Articles
Baidu academic
Similar articles in Baidu academic
[尚留记]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[尚留记]'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.