CASIA OpenIR  > 毕业生  > 博士学位论文
仿生机器鱼高机动运动控制研究
Alternative TitleLocomotion Control of Biomimetic Robotic Fish with High-Maneuverability
苏宗帅
Subtype工学博士
Thesis Advisor谭民 ; 喻俊志
2012-05-29
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水下机器人 机器鱼 机器海豚 高速、高机动运动控制 Underwater Robot Robotic Fish Robotic Dolphin High-speed And High-maneuverable Locomotion Control
Abstract仿生机器鱼及机器海豚的研究从最初的理论分析、简单功能模仿,发展到现在开始追求真实鱼类的运动性能。本文对鱼类运动性能的三个基本方面:高效率、高机动与高游速,分别进行了较为细致的讨论。根据对鱼类的观察,提炼出了适用于机器鱼运动控制的底层及上层闭环算法,不但在运动性能指标上有较大提高,而且在精度上也有较大改善。 首先,针对现有鱼类稳态游动模型中对鱼体波的解释过于简单的问题,以获得高效率的游动为前提,提出了基波的概念,建立了基于鱼体线密度的鱼体波方程,形成了较为系统的鱼体波生成方法。此外,基于该方法,进一步解释了鱼体的形态学设计,特别是鳍的位置与功能。仿真结果显示,所建立的鱼体波方程可以生成符合鱼体波要求的波形。并指出了在小振幅的前提下,著名的Lighthill 鱼体波是本文提出的鱼体波的特例。 其次,研究了鱼类瞬态运动的典型例子—— C 形起动,针对其每个阶段的不同特点分别提出了小尾模式、大尾模式以及基于虚拟 C 形管道的动态轨迹法。不但提高了弯曲阶段的角速度,而且使机器鱼亦能以最快的速度完成伸展阶段,进一步提高了 C 形起动的速度。在实验中,实现了机器鱼 670◦/s 的峰值转向角速度。另外,一次鱼体弯曲就可以实现 0 ∼ 213◦ 之间的任意转向角,从而一次 C 形起动就可以完全覆盖所有可能的方向,进一步提升了机器鱼的机动性。同时,借助陀螺仪完成的闭环算法,实现了首次转向误差不大于 10◦ 的转向精度。 再次,将动态轨迹法应用在机器海豚中,实现了精准的俯仰控制,并首次完成了在竖直平面内的前滚翻、后滚翻及其复合特技动作。对翻滚动作所建立的模型进一步说明了影响翻滚的因素,通过仿真结果与实验结果的对比,验证了模型的有效性。 最后,通过新的机构设计、选用轻质材料、以及基于尾鳍攻角的控制算法,完成了新一代高游速机器海豚的研制工作。由于该机器海豚的游速较高,使得胸鳍可以产生较大的力矩,因此设计了基于胸鳍的偏航控制算法。在较高游速的基础上,进行了机器海豚的跃水实验,以期像真实海豚一样能跃出水面。通过多次实验,最终达到了每秒1.5 倍体长的最高直线游速,以及 2/3 倍体长的最大跃水高度。
Other AbstractThe research of biomimetic robotic fish and dolphins has evolved from initial theoretical analysis to simply functional imitation, and now to the pursuit for the equivalent locomotion performance to the real fish. This dissertation provides a detailed discussion on the three fundamental aspects in the fish locomotion: high efficiency, high maneuverability and high swimming velocity. Inspired by real fish, new closed-loop control strategies both for the low-level and high-level missions are proposed. It greatly enhances the locomotion performance and offers an improvement in the locomotion precision. Firstly, in view of the over-simple explanations on the body wave employed by the existing steady swimming models, we propose a concept of “base wave”, build the body-wave function based on the linear density of the fish body, and further form a more systematic design method of the body wave with the purpose of obtaining high efficiency. By this method, the fish morphology can reasonably be explained, especially the positions and functions of fins. The simulation results verify the wave form of the proposed body wave function. Under the assumption of small amplitude, the famous Lighthill’s body wave is the special case of our proposed one. Secondly, C-start, the classic example of those transient movements of fish, is studied. Small caudal fin mode, large caudal fin mode and virtual C-shaped pipeline based dynamic trajectory tracking method are proposed for the different turning phases. It significantly improves the angular velocity in the bending phase, and also makes the robotic fish fulfill the unbending phase as soon as possible. In the experiments, a peak turning angle rate of 670◦/s is achieved. Besides, any turning angle varying from 0 to 213◦ can be implemented only by one single bending of the fish body, which assures all the possible directions can be covered in a single bending. It further enhances the maneuverability of the robotic fish. Meanwhile, a “first turning error” of 10◦ is attained by using the closed-loop control method with the aid of a gyro. Thirdly, based on the dynamic trajectory tracking method, a robotic dolphin fulfills precise pitch control, and performs 360◦ front/back flip stunts and their combined ones in the vertical plane. The dynamic model of the flip stunts f...
shelfnumXWLW1795
Other Identifier200918014628011
Language中文
Document Type学位论文
Identifierhttp://ir.ia.ac.cn/handle/173211/6447
Collection毕业生_博士学位论文
Recommended Citation
GB/T 7714
苏宗帅. 仿生机器鱼高机动运动控制研究[D]. 中国科学院自动化研究所. 中国科学院研究生院,2012.
Files in This Item:
File Name/Size DocType Version Access License
CASIA_20091801462801(4807KB) 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
[苏宗帅]'s Articles
Baidu academic
Similar articles in Baidu academic
[苏宗帅]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[苏宗帅]'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.